Skip to main content

Full text of "Yuccas of the southwest"

See other formats


Historic, archived document 

Do not assume content reflects current 
scientific knowledge, policies, or practices 



IL'3 



Yuccas 



OF THE SOUTHWEST 



JOHN MILTON WEBBER 





AGRICULTURE MONOGRAPH NO. 17 
U. S. DEPARTMENT OF AGRICULTURE 



Yuccas 



OF THE SOUTHWEST 



JOHN MILTON WEBBER 

Cytologist 

Division of Cotton and Other Fiber Crops and Diseases 

Bureau of Plant Industry, Soils, and Agricultural Engineering 

Agricultural Research Administration 




AGRICULTURE MONOGRAPH NO. 17 
U. S. DEPARTMENT OF AGRICULTURE 



Contents 



Page 

Ecology 2 

Yucca enterprises 5 

Available material 5 

Uses and potential uses of yucca 7 

Economic factors 9 

Past and future of yuccas... 10 

Yucca studies and plant characters 12 

Nature and range of study 14 

Method of presentation 15 

Sections of the genus Yucca 16 

Key to Yucca species 16 

Indehiscent-fruited yuccas 18 

Dehiscent-fruited yuccas 32 

The yucca moth, pollination, 

and fertility 65 



Page 

Yucca seeds and seedlings.,.. 67 

Germination 70 

Seedlings 71 

Root and shoot characters of 

mature yuccas 77 

Indehiscent-fruited species 77 

Dehiscent-fruited species 80 

Rhizomes produced in yuccas 84 

Recover}' of harvested yuccas 85 

Growth rates and age of yuccas 87 

Transplanting and asexual propagation.. 90 

Literature cited 93 

Index to Yucca species 95 



SOVERNMENT PRINTING OFFICEil953 



For sale by the Superintendent of Documents, Washington, D. C Price 50 cents 



YUCCAS OF THE SOUTHWEST 1 

JOHN MILTON WEBBER 2 

Cytologist 
Division of Cotton and Other Fiber Crops and Diseases 



YUCCAS are among the most valuable plants of the southwestern deserts. 
Their heads of leaves and their large stems intercept rainfall and reduce 
its flow, and the rhizomes and long, spreading lateral roots protect the soil 
from washing or blowing away. The dead yucca leaves, as well as the waste 
material of animals associated with yuccas, greatly increase the organic content 
of the soil. This waste material not only enriches the soil but materially in- 
creases its water-holding capacity. In many desert areas, each yucca, like many 
other desert shrubs, gradually improves its immediate surroundings, until a 
small community of its own is established. This community, consisting of 
small plants, rodents, insects, and occasionally birds, could not exist without 
the protection of yucca. 

Yuccas provide food, shade, and wind protection for livestock, antelope, 
deer, birds, and many rodents and other small animals. All grazing animals 
relish the young succulent leaves, and in many areas the older yucca leaves are 
almost the only permanent food that cattle have to carry them through the 
dry season. Although the pungent apex of the leaves discourages grazing, 
cows learn to chew the leaves from the stem outward and to extract the heart, 
which is very palatable, of the head of leaves. On overgrazed lands, goats 
eat the entire heads of leaves. Officials of the New Mexico Cattle Growers' 
Association stated that large herds of cattle have been maintained for 2 months 
on ranges where fresh Yucca elata leaves supplied fully half the feed and 



Submitted for publication May 16, 1952. 

2 The writer is gratefully indebted to the late Prof. Herbert John Webber, formerly 
of the University of California, for his suggestions, arguments, and assistance in the 
field and laboratory; and to Dr. Irma E. Webber, collaborator, U. S. Department of 
Agriculture, for her suggestions and advice in the preparation of this manuscript. The 
writer is also under obligation to the following research workers for acting as guides 
during his visits to various sections of the country: Robert H. Peebles, U. S. Department 
of Agriculture, Sacaton, Ariz.; Kenneth W. Parker, U. S. Department of Agriculture, 
Tucson, Ariz.; Dr. Carl B. Wolff, Rancho Santa Ana Botanic Garden, Anaheim, Calif.; 
Orville A. Parsons, U. S. Department of Agriculture, Lamar, Colo.; Dr. A. L. Hershey, 
New Mexico College of Agriculture and Mechanic Arts, and Albert R. Leding, U. S. 
Department of Agriculture, State College, N. Mex. ; D. R. Burnham, U. S. Department 
of Agriculture, Tucumcari, N. Mex. ; V. L. Cory, Southern Methodist University, Dallas, 
Tex.; and Quinton A. Hare, U. S. Department of Agriculture, Delta, Utah. Miscel- 
laneous information, seed, and fresh and dried specimens were supplied by the following: 
D. R. Burnham; R. H. Peebles; V. L. Cory; O. A. Parsons; Dr. L. M. Pultz, Univer- 
sity of Arizona, Tucson; Kenneth I. Ross, U. S. Department of the Interior, Mesa Verde 
National Park, Colo.; Fred Ares, U. S. Department of Agriculture, and Prof. C. W. 
Botkin, New Mexico College of Agriculture and Mechanic Arts, State College; Charles 
G. Marshall, U. S. Department of Agriculture, Albuquerque, N. Mex.; Bernell McAllister, 
U. S. Department of Agriculture, Price, Utah; Prof. Arthur H. Holmgren, Utah State 
Agricultural College, and Dr. F. W. Lieberman, U. S. Department of Agriculture, Logan, 
Utah; Dr. Seville Flowers, University of Utah, Salt Lake City; and Leslie N. Goodding, 
formerly of the U. S. Department of Agriculture. 

1 



2 AGR. MONOGRAPH 17, U. S. DEPT. OF AGRICULTURE 

that in many cases the ranges would suffer as much as 50 percent damage if 
the yuccas were harvested. 

Yucca has been used most extensively for its fiber. The leaves of most 
species are long and slender and furnish a fiber that is suitable for the manu- 
facture of binder twine and for rope in which great strength is not required. 
During World War I about 80 million pounds of yucca fiber from leaves 
collected in New Mexico and Texas was used as a substitute for jute in coarse 
bagging for covering cotton bales. In World War II large acreages of Y. 
glauca were harvested in New Mexico, shipped east, and made into a special 
heavy paper for the United States Navy. 

The fibrous stems of Y. brevifolia and inflorescences of Y. whipplei are used 
in the manufacture of many novelties, such as post cards, scrapbooks, book 
bindings, napkin rings, window dressings, and lightweight clubs used in play 
and comedy. At the present time veneered yucca is being widely advertised 
in southern California for interior wall covering. It is said to act as an in- 
sulating material and to improve acoustics, as well as to beautify the walls. 

ECOLOGY 

Yuccas are distributed over considerable areas in our southwestern plains, 
valleys, and mountains. They occur near sea level in southern California to 
8,000 or more feet elevation in the Rocky Mountains of Colorado. Although 
in the montane forest the yuccas are usually thinly scattered in exposed areas, 
they occasionally form small patches.- On the lower mountain slopes and 
bajadas they show a preference for southern and western exposures, and in 
these regions they usually occur in large belts. Except for Yucca elaia and 
Y. glauca, the plants of the plains and valleys are thinly scattered or in small, 
dense patches. Y. elata and Y. glauca commonly occur in belts, which extend 
for many miles. 

In the mountainous parts of central Colorado, the yuccas are subjected to 
frequent cloudiness, 20 to 40 inches of precipitation yearly, and moderately 
to extremely low temperatures. Snow has been recorded every month of the 
year in several yucca localities, and the plants are frequently covered with 
40 or more inches of snow each year. The growing season in these areas is 
90 to 140 days. In the mountainous parts of southwestern Colorado and 
Utah and in northern Arizona and New Mexico the yuccas are exposed to 
somewhat milder conditions, and in similar regions of southern New Mexico, 
Arizona, and California and southwestern Texas they are mainly subjected 
to a desert or semidesert climate. 

The yuccas of the valleys and mountain slopes of the deserts and semi- 
deserts undergo cloudless, hot days, with very little rainfall and with low 
humidity. Maximum temperatures of 95° to 120° F. are common in the 
summer, and winter temperatures rarely fall below 20° F. Although the 
annual precipitation varies from approximately 5 to 20 inches, very little of 
the area receives more than 10 inches of rainfall. In southern California the 
rainfall is mainly limited to the winter months, but in southwestern Texas, 
New Mexico, and eastern Arizona summer rainfall predominates. The growing 
season in these desert areas is from 195 to 240 days. 

On the Edwards Plateau, west-central Texas, the yuccas are exposed to 
semiarid conditions. The rainfall is about 25 inches, and the temperatures 
range from near zero in the winter to 110° F. in the summer. The atmospheric 
humidity is rather high, and the growing season averages about 230 days. 



YUCCAS OF THE SOUTHWEST 3 

Southeast of the Edwards Plateau the yuccas are found in a zone of transition 
from the deserts of the west to the humid areas of the east. West of the 
Edwards Plateau the desert conditions become increasingly severe, until they 
reach their maximum in southern Arizona and eastern California. The semi- 
desert conditions of the Plateau extend north and northwest into the plains 
of northwestern Texas, southeastern Colorado, and northern New Mexico and 
Arizona. In the latter sections, however, the rainfall is infrequently distributed 
and somewhat lower. The yuccas in southeastern Nevada, west-central and 
eastern Utah, and northwestern Arizona are exposed to, rather severe desert 
conditions, whereas those south and west of the Mojave Desert of California 
grow in a mild desert climate. 

Although yuccas are found in all types of soil, they are more common in 
sands and gravels. Y. baileyi, Y . angustissima, Y. constricta, and especially 
Y. elata and Y. glauca are particularly common on the more compact sandy 
soils. These sands are occasionally red, usually 4 or more feet deep, and fre- 
quently underlain with caliche or granite hardpan. The remaining species 
exhibit a definite preference for a rather coarse-textured gravelly loam. These 
soils are usually 2 or 3 feet deep and are mainly underlain by a rock bench. 
The former sandy soils are characteristic of desert grasslands and western 
prairies, while the latter are commonly associated with desert plains, alluvial 
fans, and mountain slopes. 

Yucca treculeana, Y. torreyi, Y. schidigera, Y. arizonica, Y. brevifolia, 
Y. wbipplei, and Y. ro strata are chiefly found on gravelly soils. Although 
Y. thompsoniana, Y. rupicola, Y. reverchoni, Y. neomexicana, and Y. gil- 
bertiana are also gravelly soil plants, they frequently occur on rocky ledges 
and mesas, where the soil is very shallow. On the ledges the plants grow in 
the gravel-filled fissures of the rock bench. Y. rupicola appears to prefer a 
limestone soil, but many fine plants of Y. gilbertiana occur in a rather finely 
decomposed slate soil. Y. camerosana and, especially, Y. faxoniana exhibit 
a preference for the coarse gravelly and rocky soils of the mountain slopes and 
mesas, whereas Y. schottii mainly occurs in gravelly soils of mountain ravines. 
Although Y. baccata is mainly a woodland, gravelly soil plant, it occasionally 
extends into the compact sands of the grasslands. In California Y. ivhipplei 
is rather common in red clay, and in southwestern Texas Y. torreyi frequently 
occurs on a similar, poorly vegetated soil. On the eastern slopes of the Edwards 
Plateau, Tex., Y. rupicola is occasionally found in a calcareous clay, and 6 
miles south of Kiowa, Colo., Y. glauca occurs in a rather large clay marsh, 
which appears to be quite saline. The yuccas in the latter area are extremely 
vigorous, and they do not appear to be suffering from excess moisture or salt. 
They differ from typical Y. glauca in the following characters: The plants are 
extremely dense, having 33 or more heads of leaves per square meter; the 
leaves are very glaucous, flexible, and sparsely filiferous; and, although the 
leaf apex is sharply pointed, the point is flaccid and nonpiercing. 

All of the preceding soils, except those in the Kiowa locality, are well 
drained and have little or no subsurface water. It is a common expression 
in the desert regions that, "Yuccas do not like to have wet feet." Wolf (34)* 
has reported that excess irrigation of Y. schidigera caused heart rot. The 
disease is common in other species of yuccas if they are overirrigated. All 
species in the writer's nursery, except Y. rupicola and possibly Y. reverchoni, 
are more vigorous than native plants. This extra vigor is undoubtedly the 



3 Italic numbers in parentheses refer to Literature Cited, p. 93. 



4 AGR. MONOGRAPH 17, U. S. DEPT. OF AGRICULTURE 

result of an increase in water, coupled to some extent with better and richer 
soil, and, possibly in a few cases, with a longer growing season. Native plants 
of Y. rupicola are undoubtedly subjected to more rainfall than those in the 
writer's nursery. 

The greatest response to water is that of length of life of the leaves, which 
is reflected in the size and shape of the heads of leaves. When rainfall is 
insufficient the outer leaves of the head die, and in severe drought periods 
only those leaves imbricated around the buds may remain fresh. On the other 
hand, with sufficient water the leaves remain alive considerably longer than 
usual and the leaf heads are exceptionally large, or even wanting in cases 
where the leaves remain fresh along the entire length of the stem. The size 
of the leaves is probably equally as responsive to variations in supply of water. 
In the Rancho Santa Ana Botanic Garden, the average length of leaves of 
unirrigated Y. schidigera is 67.5 cm., while that of irrigated plants is 80.1 cm. 
Although the leaves of the irrigated plants are somewhat broader than those 
of the unirrigated ones, no measurements have been made. Yuccas with extra 
large leaves and leaf heads are common in gardens, along roadsides, or in 
fields where cattle have worn trenches around the bases of the plants. 

Undoubtedly stem growth and sprout formation are correlated with leaf 
size and longevity. In the Y. schidigera plants discussed above, the unirrigated 
plants average 1.62 sprouts per plant and their main stems average 41 cm. tall. 
The irrigated plants average 3.93 sprouts per plant and their main stems ap- 
proximately 85 cm. tall. Furthermore, many of the sprouts of the irrigated 
plants are nearly as tall as the mother shoot, but only a few of the sprouts 
of the unirrigated plants have developed stems of appreciable length. In these 
yuccas, the stems of all the irrigated plants are entirely covered with fresh, 
green leaves, whereas, except for a normal-size apical crown, the stems of the 
unirrigated plants are covered with dried, reflexed leaves. The writer has found 
a similar increase in sprouting of semi-irrigated plants of Y. baccata, Y. torreyi, 
and Y. schottii. 

Yuccas are associated with many vegetational communities and plants. In 
the desert and plain grasslands they are commonly associated with the follow- 
ing perennial grasses: Blue and hairy grama {Bouteloua spp.) ; three-awn 
{Aristida arizonicd) ; galleta {Hilaria spp.), tobosa (Hilaria mutica), and 
curly mesquite {Hilaria belangeri); dropseed {Sporobolus spp.); and beard- 
grass {Kndropogon spp.). The common shrubs of these regions are fourwing 
and spiny saltbush {Atriplex spp. ) , velvet and honey mesquite {Prosopis spp. ) , 
rabbitbrush {Chrysothamnus nauseosus), and snakeweeds {Gutierrezia spp.). 
In the more sandy and higher elevated grassland areas, the yuccas are com- 
monly associated with Indian ricegrass, or sand bunchgrass (Oryzopsis hy- 
menoides), Navajo ephedra {Ephedra viridis var. viscida), shinnery oak 
{Quercus havardi), and beargrasses {Nolina spp.). 

On the lower mountain slopes, bajadas, and mesas, the yuccas are com- 
monly associated with such shrubs as coastal sage {Salvia sp.), chamise 
{Adeno stoma fasciculatum), creosotebush {Larrea tridentatd) , Jatropha cuneata, 
ocotillo {Fouquieria splendens), blackbrush {Coleogyne sp.), beargrasses 
{Nolina spp.), sotols {Dasylirion spp.), sumac and poison ivy {Rhus spp.), 
catclaw {Mimosa biuncifera), Torrey ephedra {Ephedra torreyana), agritos 
{Berberis tri folio lata) , snakeweeds (matchweeds) {Gutierrezia spp.), and 
Lycium spp. Many cacti and several agaves are also found admixed and asso- 
ciated with the yuccas of these regions. In the desert woodlands the yuccas 



YUCCAS OF THE SOUTHWEST 5 

are also commonly associated with several species of oak (Quercus spp.), 
juniper (Juniperus spp.), and pinyon (Pinus cembroides). 

In the montane forest, the yuccas are commonly associated with pines, 
spruces, and firs, as well as several broadleaved trees and many shrubs. 

Trelease (31, p. 123) called attention to the fact that yuccas are characteris- 
tically xerophytes and stated, "Like other arborous Liliaceae, their larger repre- 
sentatives produce the impression of being the culmination of a vegetative type 
perhaps formerly of wide distribution, but now barely able to hold its own 
except in desert regions where competition between plants is less than else- 
where, while structural adaptation enables them to endure the rigors of this 
last resort." That yuccas are well adapted to desert conditions is unquestion- 
able. Their large stems and rhizomes store considerable moisture, and, as 
Trelease pointed out, ". . . their leaves are well guarded against undue trans- 
piration." Relative to the seeds of the fleshy-fruited yuccas, Trelease stated, 
". . . their abundant endosperm suggests an adaptation to the dry regions." 
Although the seeds are well protected, supplied with abundant food, and retain 
their viability under natural conditions for several years, the writer has found 
that they require considerable moisture for germination. In many of the desert 
regions the rainfall is insufficient to germinate yucca seeds and, hence, the 
plant's reproduction is considerably diminished. 

Yucca seedlings, as well as mature yuccas, are extremely persistent and well 
able to compete with the majority of their associates. This persistency is shown 
by the failure of one plant to penetrate the grounds occupied by another plant. 
It is questionable, however, if they can long endure the shade of the taller 
desert shrubs and trees, and it is doubtful if any plant can compete success- 
fully with shinnery oak, and possibly several mesquites. The shinnery, or shin, 
oak is one of the most remarkable plants of our southwestern semideserts. 
On the Llano Estacado of western Texas it covers immense areas, often in 
pure, or almost pure, stands, and rarely exceeds 70 to 90 cm. in height. In 
the sandy regions near Plains, Yoakum County, Tex., shinnery oak occurs in 
stands that are a quarter or more mile square. The areas surrounding these 
stands are well covered with grasses and small annuals and perennials, includ- 
ing Y. glauca, but none of these plants intrude to any extent in the oak's 
domain. In fact, the stands are so dense that one hesitates to enter them in 
fear of hidden rattlesnakes. According to V. L. Cory, 4 the huge area over 
which the shinnery oak grows has been referred to as "the largest oak forest 
in the world." Similar competition between mesquite and Y. glauca occurs 
near Spur, Tex. The growth of the mesquite, however, is not so dense as that 
of the shinnery oak, and Y. glauca, as well as other perennials, are somewhat 
admixed with the mesquite. 

YUCCA ENTERPRISES 

Available Material 

During World War II the writer conducted a survey to determine the 
amount of raw material that might be harvested from the denser concentrations 
of southwestern yuccas. The areas of these concentrations were determined by 
cyclometer readings, in conjunction with observations, contours, elevations, 
and distances calculated from maps. The density of the stands and weight 



4 In correspondence, dated Aug. 20, 1945. Mr. Cory was formerly range botanist, 
Texas Agr. Expt. Sta., Sonora, Tex., and is now field botanist, Southern Nfethodist 
University, Dallas, Tex. 



AGR. MONOGRAPH 17, U. S. DEPT. OF AGRICULTURE 



.2 « 
o-~ 



>« o 



-. CO 



O COOt* 

O 0O<©** 

^ CO I>^CO CN 

g jo of to~cT 



OcO 

00 o 

ooo 

H OS 



NOONO0CO 

■<* ts of © co~ *o of 



OO <N 

CO 00 CO 

co~to~ 1> 

tO CO 



000000:000 0<M 

OO'MTfMXHOO CO— i 

^XXiC-<tNlNM© CO O 

§ oo^i-Teo cOi-ToTt> ocT^T 

t, HlCH CO lO 



CN O) 00OO Ol O 
MCONCOO©^ 

o^co^c^cooq^co^e* 

rjT oT co~ i>~ oo" r-T of 

tF tO •**< CO Tt* 



o coooc 

Oi O) C5 CO 

cx^oq^-^co 

o~co~oo"'oo"' 

CO OOCN 



2 ° 



8 £ 

t-. o 



oo 



»o to 

x oo 



CO to CO Tfi CO 



CO 
OJCO 






o to 

-^ oo oo oo to oo oo 
i-i . * - " *© " 

r^cOCSI -tf CN 



O 

bD^3 do 



C5 © © © 



O O 3 



11 



V CJ 
•<s> "«a 

« § § « 



C C3 

C3s Cs, 



S3 



§53 1 5 

§J3J3J3 O 



e e 






-O 



* 






3 


<D,C 


bC b£ 


c3 


c 


> 


is 


ri 


+3 


0) 


u 






ri 


CJ 



Q> © « 

511 






rH 00 0)0 
tO rf CN O CM 



o tO to to 
00 CN to 00 



5fi 




C CO 

CD c3 
£^ 
ftjj 

•2 © 

0Q*O. 

2 S 

3 O 

S5 



YUCCAS OF THE SOUTHWEST 7 

of raw material were determined by actual counts and weights of harvested 
material from several half-acre strips scattered throughout the concentration. 
It was assumed that no area bearing less than one-fifth ton of fresh leaves 
per acre could be harvested profitably, and all such areas were deducted from 
the concentrations. Table 1 gives a summary of the principal native stands 
of yucca in the southwest, with estimates of yields. 5 

The possible yields of several other areas bearing dense concentrations of 
yucca have not been determined. The largest of these areas is that of Yucca 
brevijolia var. jaegeriana surrounding Cima, Calif. Unquestionably, this is 
the largest and densest yucca concentration in the Southwest. Other large 
concentrations observed by the writer are as follows: 

Yucca brevijolia, south of Boran, and between Hesperia and Morango Valley, Calif. 
Y. brevijolia var. jaegeriana. White Hill district, Mohave County, Ariz. 
Y. arizonica, Fresnal Canyon, Baboquivari Mountains, Ariz. 

Y. glauca, Railroad Mountain, near Elkins, N. Mex., and vicinities of Calhan, 
Kiowa, and Monument, Colo. 

Reported concentrations not observed by the writer are: 

Yucca brevijolia, 12. miles west of Congress Junction, Ariz. 

Y. brevijolia var. jaegeriana, between Littlefield and" Grand Wash, Ariz. 

Y. glauca, near Chillicothe and Sedgwick, Colo. 

Reported dense concentrations of Y. glauca near Spur and Lubbock, Tex. 
(18) were not found by the writer. It is believed that these stands have been 
taken over by shinnery oak and mesquite or that the lands have been cleared 
for agricultural development. 

Uses and Potential Uses of Yucca 

Yuccas had an important place in the early southwestern Indian and Mexican 
cultures. Fiber and twine, identified as made from yucca and determined to 
be around 2,000 years old, was found in Canyon del Muerto ruins, Arizona, 
and Lapham (21) reported that beams and timber of Yucca brevijolia have 
been found in ancient cliff dwellings. Early and present-day uses of yuccas 
for food, beverages, detergents, medicines, clothing, construction of dwellings, 
and household articles by the Indians and Mexicans were reported by Trelease 
(31), Standley (29), Yanovsky (37), Kearney and Peebles (20), and others. 
In the opinion of Bell and Castetter (2) ". . . yucca ranked foremost among 
the wild plants utilized by the inhabitants of the Southwest. It holds this place 
because of the great variety of uses to which it could be put and to the wide 
accessibility of the genus within the Southwest." 

Probably the first large-scale utilization of yucca in the United States was 
the use of Y. elata and Y. glauca to supplement cattle feed during the drought 
from 1916 to 1919. Forsling (14), Wooton (36), Jones and Conner (18), 
and others reported that the dead leaves were burned off the yuccas and the 
plants cut and chopped for silage. The chopped yuccas, supplemented with 
such concentrates as cottonseed meal, were found to be a practical means of 
maintaining cattle during severe droughts. 



5 See typewritten report filed in the Division of Cotton and Other Fiber Crops and 
Diseases, Plant Industry Station, Beltsville, Md.: Webber J. M. yucca stands and 

DISTRICTS, AND ESTIMATED YIELDS, WITH MISCELLANEOUS NOTES ON HARVESTING, 

fiber and costs. 76 pp., illus. 1943. Revised 1944. 

See also mimeographed report: Robinson, B. B., and Webber, J. M. yucca. Bur. 
Plant Indus., Soils, and Agr. Engin. 6 pp. 1948. 



8 AGR. MONOGRAPH 17, U. S. DEPT. OF AGRICULTURE 

According to Cruse (11 J, the chemical composition of yucca on a dry- 
weight basis was 4.4 percent protein, 6.7 percent ash, 1.5 percent fat, 3.2 
percent fiber, and 49.2 percent carbohydrates. 

Undoubtedly the most extensive use of yucca has been for the production 
of fiber. According to Botkin and associates (4, 5, 6) the leaves contained 
approximately 50 percent moisture and about 40 percent chemically extractable 
fiber. These authors found that the strengths of Y. glauca and Y. elata fiber 
were, respectively, about 45 and 47 km., as compared with 33 km. for jute; 43 
for palma istle and African sisal; 47 for Wisconsin hemp; and 62 for Manila 
hemp (abaca). In baccate-fruited species, they reported the following aver- 
ages, in kilometers: Y. baccata, 60.2; Y. torreyi, 61.1; Y. schidigera 50.4; 
and Y. faxoniana, 54.0. Botkin and Shires concluded that the fibers of Y. 
glauca and Y. elata were suitable for making twine and burlap, while the 
fibers of the other tested species were strong enough for making binder twine 
and ordinary rope. It is generally considered, however, that yucca fibers do 
not possess values that would make it a strategic material, except in extreme 
emergenq'. 6 

Within recent years yucca juice has come to be considered of greater value 
than the fiber. Its utilization as a base in liquid fertilizers is said (1) to re- 
duce surface tension of irrigation water, greatly assisting penetration in heavy 
soils; (2) to assist in soil flocculation to a marked degree; and (3) to serve 
as a carrying agent for the plant-food chemicals. Relative to this use Haynes 7 
stated, "... yucca extract, a liquid organic acid . . . has proven to be the best 
medium known to date for quick assimilation. The yucca extract itself is rich 
in the vital minor elements including boron, iron, magnesium, manganese, 
copper and zinc." 

The juice is said to be widely used as a carbon dioxide stabilizer in the 
control of oil fires, and saponin from yucca is considered a good base for 
soaps, shampoos, cleansing powders, and tooth pastes and powders. 

The rhizomes of yucca are reported to be a rich source of a female hormone, 
desorycorticorcerone, which is of great value in gynecological work. 

Katz, Hall, and Petersen (19) suggested that Y. brevifolia, particularly the 
wood from which the water-soluble constituents have been extracted, might 
prove a source for lignaceous materials, possibly valuable in synthesis of 
vanillin. 

In a treatise on California wild tree crops that possibly could be utilized 
during emergency periods, as well as under normal conditions, Wolf (35) 
reported that the seeds and dry pods of the Joshua-tree (Y, brevifolia) are 
fairly good feed materials. His analysis of the pod material showed that it 
compared very favorably with dried orange peel. Wolf's figures for proximate 
percentages of chemical composition of harvested pods in comparison with 
that (in parentheses) of dried orange peel, were as follows: Water, 7.6; 
protein, 6.7 (5.8); fats, 2.0 (0.7); fiber 16.8 (10.6); nitrogen-free extract 
(carbohydrates) 60.0 (64.7) ; and ash (mineral matter) 6.9. He reported 
the seed of the Joshua- tree to contain 34.4 percent oil, as compared with 36.4 
percent for flaxseed, 21 to 23 percent for cottonseed, and 18 percent for soy- 
beans. The seed meal, or residue left after oil extraction, was roughly equiv- 
alent to wheat bran (in parentheses): water, 9 percent; crude protein, 15.5 



6 See footnote 5, p. 7. 

7 Haynes, J. D. a newer method of liquid fertilization. 7 pp. n.d. Liqui- 
nox Co., Orange, Calif. [Processed.] 



YUCCAS OF THE SOUTHWEST 9 

(15.8) percent; fat, 5 (5) percent; fiber 15 (9.5) percent; nitrogen-free 
extract 53 (54.3) percent; and ash, 2.5 (6.0) percent. Wolf (35, p. 15) 
stated, ". . . the Joshua Tree oil is a clear, light yellow oil, practically taste- 
less and odorless. Its iodine number of 120.3 indicates that it is a semi-drying 
oil which along with its other characteristics indicates that it may be found 
to have some value for other uses than merely as an edible oil for salads, 
shortening, etc." 

Cruse (11) stated that several authors have reported fructose, fructosans, 
dilevans, and glucose present in several yucca species. The fruits of the baccate 
species were very sweet, and when they were drying in an oven they gave 
off a very sweet candylike odor. A single fruit of Y. baccata or Y. arizonica 
will often fill a quart jar. 

Economic Factors 

In order that the potential values of yucca may be thoroughly understood, 
further investigations should be conducted. Under normal conditions, how- 
ever, it is doubtful if native yuccas could be economically used for large-scale 
projects in the United States. Even the use of yuccas in a combination of two 
or three of the most promising enterprises must be considered speculative at 
the present time. Although in the majority of yucca concentrations there are 
very dense areas, the ordinary or average stand would yield only approximately 
0.55 ton of fresh leaves, or 0.85 ton of fresh crowns per acre. Reliable in- 
formation indicates that the rate of harvesting and assembling of the material 
will not greatly exceed 500 pounds of leaves or crowns and 2,000 pounds of 
stems with crowns per day per workman. The transportation of raw material 
to a factory is one of the major expenditures. Although in some localities 
the terrain may permit trucks to move with little difficulty, in others sand, 
gull*es, and mountains would prohibit trucking or make it very costly. 

The present experimental data as to quantity, quality, or composition of 
yucca fiber, extract, and hormones are not sufficient to determine what would 
be obtained or expected in large-scale operations. The common reductions or 
losses in large-scale operations, in conjunction with excessive costs of collect- 
ing and trucking native yucca material, indicate that the majority of the pro- 
posed yucca projects would allow a very narrow margin of profit at the best. 
There is also some opposition to large-scale yucca harvesting. In areas where 
yuccas fail to reproduce readily or in the drier areas where the plant's natural 
values are extremely important to the land and to the ranchers, the yuccas 
should not be harvested except for use in extreme emergencies. 

To overcome many of the economic difficulties involved in the utilization 
of native yuccas, it has frequently been suggested that the plants should be 
cultivated. The culture of yuccas would not only eliminate some of the diffi- 
culties involved in the use of native plants, but very likely would greatly im- 
prove the plant. There is sufficient evidence to indicate that yuccas will grow 
very well on much of our poorer agricultural lands. If such plantings were 
cultivated to conserve rainfall or lightly irrigated, the plants would be con- 
siderably more vigorous. Undoubtedly the greatest benefit of growing yuccas 
would result from the selection of native plants most suitable for economic 
use. The writer's studies indicate that through simple selection a field of 
better and more uniform plants than occur in nature would not be difficult to 
establish. 

Although yuccas are quite free from diseases, in several areas they are badly 



10 AGR. MONOGRAPH 17. U. S. DEPT. OF AGRICULTURE 

damaged by insects and rodents. It is not known whether these infestations 
could be controlled or whether they would increase in cultivated yucca fields. 
If yuccas were cultivated, it is likely that unforseen problems and difficulties 
would develop. 

PAST AND FUTURE OF YUCCAS 

In referring to the baccate- fruited yuccas, Trelease (31, p. 123) stated, 
"... that they have been derived from the thin-seeded capsular species seems 
more probable than the reverse ..." It appears that Trelease was led to this 
conclusion by the fact that the seeds of the fleshy-fruited yuccas are better 
adapted to desert conditions than those of the capsular- fruited species. In the 
present monograph, Trelease's conclusions are confirmed by the fact that the 
fibrous- rooted (baccate) species are more characteristic of the drier sections, 
while the rhizomatous (capsular) species are characteristic of the more humid 
sections. 

Unquestionably several species of yucca are the oldest or among the oldest 
living plants of the deserts. Since the discovery by Laudermilk and Munz 
(22, 23) of yucca remains in the dung of the extinct ground sloth (Nothro- 
therium) there has been little doubt that yuccas were formerly more widely dis- 
tributed. These authors and Dr. Arthur D. Howard found that the ground sloth 
fed mainly on the Joshua-tree, and by comparisons of the plant remains of 
the dung with existing floras they concluded that the climate in the area the 
sloth inhabited was distinctly more humid than today. These findings and 
conclusions and the large, grotesque appearance of several yuccas have at- 
tracted much attention; in fact, they are not uncommonly depicted in recon- 
structions of prehistoric, mesophytic forests. Within recent years, several pic- 
tures of immense ground sloths uprooting and completely devouring huge 
Joshua-trees have appeared. These paintings are undoubtedly deceptive, as it 
is very likely that all yuccas of the past, including the Joshua-tree, were of 
a rhizomatous type, forming large, dense clumps, and of comparatively low 
stature. 

Regardless of the present wide distribution and large concentrations of 
yucca, its future appears very dim. This gloomy outlook is mainly the result 
of the plant's failure to reproduce and its destruction by man. Throughout the 
Southwest yucca seedlings are very rare and mainly limited to the more humid 
sections. In 1918 Wooton (36, p. 13) observed the lack of yucca seedlings 
and concluded, "... reproduction from seed, while possible, is but remotely 
probable in the open country. ... It thus appears that most of the species here 
listed, if used up, are not likely to return." 8 By the hands of man, yucca has 
suffered greatly through agriculture, various enterprises, fire, and wanton 
destruction. In 1918 Jones and Conner (18) reported large concentrations of 
Yucca glauca in the vicinities of Lubbock and Spur, Tex. At the present time, 
in these localities, as well as elsewhere in the more livable regions of the South- 
west, large tracts containing yucca have been cleared for necessary agriculture. 
V. L. Cory 9 stated, "The capsular species of yucca very likely are on the decrease 
in all parts of west Texas, especially where livestock are being grazed. These 
plants are of low growth, and the inflorescence shoots are highly palatable to 

8 Wooton lists Yucca elata, Y. glauca, Y. torreyi (macrocarpa) , Y. baccata, Y . brevifolia, 
Y. rupicolA, and Y . faxoniana (Samuela). 

9 See footnote 4, p. 5. 



YUCCAS OF THIi SOUTHWEST 11 

all types of livestock. It is rare for any of them to get into flower, and even 
more rare for them to produce seed. In the passage of time, the old plants 
die and there is no reproduction." In many sections in southern Texas the 
writer has seen areas so heavily grazed by sheep or goats that yuccas and 
other plants considered as emergency stock feed have been exterminated. 

With reference to the preservation of the Joshua-tree in California, Munz 
(27 J stated, "... large tracts of Joshua land have been cleared by homesteaders, 
many of whom have later had to abandon their clearings. It might be expected 
that the open vegetation of the desert would be free from devastation by fire, 
but for many miles along the highway between Victorville and Cajon Pass, 
Calif., there is a burned area of long-leaved Joshua." 

The writer has seen large burned areas of Y. scbidigera, Y. elata, Y. glauca, 
and Y. rostrata. Although in many cases such burned yuccas will recover, in 
others they are so badly burned or the area is so heavily infested with rodents 1 (> 
that recovery is impossible. 

Under the introductory paragraphs and the section on Yucca Enterprises, 
it has been shown that yuccas have been utilized in many ways. These projects 
have involved practically all yucca species, and several of them have neces- 
sitated the cutting of enormous quantities of yuccas over vast areas. Although 
the best information indicates that the majority of harvested plants will sprout 
and within a few years renew the crop, this evidence is meager and limited to 
only a few species. Furthermore, some evidence and observations indicate 
that under certain conditions these few species may not recover following 
harvesting. In reference to yuccas as stock feed, Wooton (36, p. 14 J stated, 
"It should be clearly understood that the supply of this emergency feed is not 
by any means inexhaustible; in fact, on many ranches it is scarce or very limited 
in amount; also that natural renewal is slow with the best species and im- 
probable with others, while it is reasonably rapid with but one species (Yucca 
glauca)." 

Many yuccas are damaged by breaking off parts for decorative material, 
and many others are moved to home gardens where they lack proper care and 
soon die. There have been frequent reports of Joshua-trees being wantonly 
burned and hacked, and the writer has observed many fine specimens of other 
trunklike species mutilated in various ways by tourists and campers. 

Undoubtedly self-sterility and cross-pollination add to the insecurity of 
several species. Y. thompsoniana and Y. revercboni, as well as several other 
species, are comparatively scarce, are quite self -sterile, and occur in areas 
dominated by other species, with which they unquestionably hybridize quite 
freely. If these comparatively self-sterile plants depended entirely upon seed 
for reproduction, it is very probable that they soon would be eliminated and 
replaced by the more abundant species and by hybrid derivatives. In many parts 
of the Southwest the confused nature of the yuccas is possibly indicative of 
such species conversions. 

With reference to the large forest of Y . brevifolia var. jaegeriana in Cali- 
fornia, Munz (27 ) stated, "Early action is necessary, if we are to preserve an 



10 In all of the drier parts of the Southwest rodent damage to yuccas is common. Al- 
though both rabbits and wood rats often destroy young growth, wood rats are particular!) 
damaging. These rats not only eat the flowers, fruit, and leaves, but in badly infested 
areas they frequently destroy the entire head of leaves and even gnaw into the trunk- 
like stems. In heavily infested areas from 30 to 40, and even as high as 65 percent, of 
the yuccas are badly damaged. Such infestations are quite common in Y. schidigera 
and Y. elate. Areas of damage are usually from one-fourth to several square miles in area. 



12 AGR. MONOGRAPH 17. U. S. DEPT. OF AGRICULTURE 

outstanding example of so strange and beautiful a forest ... If future gen- 
erations are to have any adequate concept of the original vegetation and other 
desert life in an undisturbed condition, and if the short-leaved Joshua is to 
be maintained at all, an area large enough to assure its protection should be 
set aside." Undoubtedly, such protection would prolong the existence of 
yuccas and it should be provided for the majority of the species. 

YUCCA STUDIES AND PLANT CHARACTERS 

In 1902 William Trelease's monograph (31) on yuccas was published. 
This work represented studies begun some 10 years previously, and up to 
1911 it was supplemented by short articles. Despite the difficulties involved 
in yucca studies and the hardships of desert travel at the beginning of the 
present century, William Trelease's accomplishments were of a high magni- 
tude, and his ability to recognize distinct forms or species with scanty material 
seems almost incredible. In several instances he separated well-defined species 
in localities where the plants were highly variable, where species intergraded 
into each other, and where the typical species were all but lacking. In the 
majority of these cases, however, with' today's known range, pure concentra- 
tions of the type have been found, possibly a hundred or even a thousand 
miles away from the type locality. 

Although Trelease occasionally mentions instability within yuccas, it is 
apparent that he did not know how extensively it occurs. Furthermore, some 
of his descriptions indicate that he failed to consider whether variations were 
caused by environmental causes or by genetical differences. As a result of these 
failures, a few of Trelease's descriptions do not include common variations 
caused by physiological reactions; others are too broad and include apparent 
hybrids. That these failures were mainly the result of Trelease's inability to 
cover a wide range is indicated by the fact that in several instances his key 
character to a species is prevalent in only a relatively small proportion of the 
present known range of the species. 

In 1938 and 1947, the Arnold Arboretum of Harvard University published 
parts one and two, respectively, of Susan Delano McKelvey's treatise (24, 25) 
entitled "Yuccas of the Southwestern United States." This publication con- 
tains excellent references to the historical nomenclature and morphological 
analyses of yuccas. In it are cited numerous collections and herbarium specimens. 
In reviewing the literature, McKelvey mentions that Engelmann and other 
early investigators had difficulty in finding specific characters and considered 
yuccas unstable. McKelvey confirmed these difficulties and concluded (24, p. 6) 
that the inconstancy "makes it obvious why any characterization of groups or 
of a species must be understood to include reservations as to exceptional 
plants," and (25, p. 7 ) that, "It should be realized that, even in important 
characters, instability is to be expected in every Yucca species — one acquires 
a 'sense' for reasonable as against unreasonable variations." 

Throughout McKelvey's yucca treatise it is clear that she believes hybridiza- 
tion largely responsible for instability. In several instances she considers 
named types hybrids, and in her Discussion of Characters (25) stated, "The 
moth which works apparently not only between plants of one but of several 
species when these come into proximity is, over the years, doubtless respon- 
sible for many individual peculiarities." Despite this recognition of hybridiza- 
tion, McKelvey segregates several new species that are extremely variable, 



YUCCAS OF THE SOUTHWEST 1 3 

that intergrade into two or more types, and that exhibit no characteristics 
typical only of the new forms. Although McKelvey commonly associates 
teratological variations with possible yucca hybrids, such variations are not 
usually indicative of hybrids. 

McKelvey (23, p. 8) commented on the characters that were useful and 
those that were of little value in determining yucca identities. Relative to the 
leaf, she stated, "The texture of the mature blade, whether grass-like and 
rather soft to somewhat flaccid, or wiry and resilient, is one of the characters 
distinguishing series in Chaenocarpa." Four of McKelvey's six series in 
Chaenocarpa are mainly differentiated by leaf texture. The writer's field and 
nursery comparisons of leaves of thousands of plants in these series indicate 
that the series and the majority of species in them cannot be separated on the 
basis of leaf texture. Throughout the present monograph it is shown that leaf 
texture is extremely variable and dependent upon environmental conditions 
and genetical composition. 

McKelvey also found the color and texture of the capsules useful in de- 
termining identities. Although the writer believes these characters to be more 
stable than those of the leaves, they are nevertheless quite variable. In several 
species the capsules are commonly white to dark brown, smooth to rough, and 
thin- to thick-walled. Although it is not known what causes these variations, 
considerable evidence indicates that color and possibly smoothness depend 
upon environmental conditions. In general, it appears that if the plant is in 
an open sandy area the capsules are lighter colored and smoother than those 
of plants in rocky or gravelly mountain areas. 

Relative to whether the capsules are constricted or symmetrical McKelvey 
(23, p. 11) stated, "Some capsules show a tendency to constriction, possibly 
not always caused by the work of the yucca moth; those with a thin wall seem 
to be more subject to it than those with a thick wall — it is not a dependable 
character." Although evidence indicates that the work of the yucca moth is 
responsible for capsular constriction, the writer believes that it has little to do 
with the thickness of the capsule wall. Possibly the capsules with the thinnest 
walls the writer has seen occur on Yucca elata plants near Aguila, Ariz. None 
of the capsules of these plants exhibited constrictions. Constricted capsules 
occur frequently in species having small capsules, and rarely, jtf ever, in pure 
species having large capsules. Unquestionably the constrictions in both the 
capsular- and baccate-fruited yuccas are mainly caused by the failure of the 
ovules to develop in the region of the ovary pierced by the moth's ovipositor. 
It is very likely that the failure of the ovules to develop depends upon how 
deeply the ovary is pierced, which probably depends upon the diameter of the 
ovary and the size of the placenta as well as the thickness of the wall, and 
possibly such features as length of style and shape of flow T er. Regardless of the 
causes of fruit constriction, outside of hybrid zones, the presence of the char- 
acter or the lack of it appears to be quite constant. 

It should be realized that many of the morphological problems and growth 
peculiarities of yuccas can be solved and evaluated only when the plants are 
grown under uniform conditions and are thus subjected to comparison and ex- 
perimentation. Such studies, however, have been greatly handicapped by the 
extremely slow growth and development of the plant and by its large size. 

In the present monograph, evidence is presented that indicates variations in 
yuccas are mainly of three types: (1) Environmental; (2) minor genetical; 
and (3) major genetical. Variations caused by environmental and minor genet- 
ical differences are to be expected and are not out of line with species or 



14 AGR. MONOGRAPH 17, U. S. DEPT. OF AGRICULTURE 

form characterizations. On the other hand, variations owing to major genet- 
ical differences are not in line with species or form characterizations, and 
plants showing such variations appear either (1) to possess specific characters 
of two or more species or (2) to be major mutants in one or two specific 
characters of a species. Plants exhibiting mutation are rare and may possibly 
be considered "exceptional" plants of a particular type. They do not exhibit 
specific characters of species or forms other than those of the group to which 
they belong. 

Variants that exhibit characteristics of two or more species or forms appear 
to be intertype hybrids and recombinations. Although phenotypically these 
plants are frequently midway between species, they are more often gradients 
between the apparent types involved. Such gradients often form large transi- 
tional zones between the ranges of various yucca species. The plants within 
such a zone usually exhibit a gradation from one type to another, and occa- 
sionally colonies or even large patches of fairly uniform plants appear to have 
been set aside and become well established. As the plants in these patches 
can readily be characterized and appear rather stable, they are easily segregated 
as types. Nevertheless, they fit into a gradation and exhibit specific or slightly 
modified specific characters of the species or form involved in the gradation. 
To separate such patches into named types only increases the taxonomic 
perplexity of yuccas. 

In order to minimize the confusion, it appears that species scopes should 
be broadened to include minor differences and that those plants which are 
combinations of well-defined types should be recognized as hybrids or transi- 
tional forms. To identify satisfactorily all transitional types or forms would 
be a tedious undertaking and without great value. 

NATURE AND RANGE OF STUDY 

The present botanical study was a part of an 8-year survey of the yuccas of 
the southwestern United States conducted to evaluate the plants as a possible 
source of cordage fiber. Although the botanical work is academic in nature, 
many of the data presented have a direct bearing upon the possible commercial 
utilization of the yuccas. 11 

The botanical part of the work was carried out primarily to overcome the 
taxonomic confusion mentioned above and to determine the cause of the pe- 
culiar growth behavior of the yuccas. In the accomplishment of these objects, 
it was hoped that practically all yuccas in the Southwest could be identified 
and that many of the problems relative to the growth of the plant would be 
clarified. One of the main interests in the study was concerned with the 
problems that deal with the establishment of variations, their perpetuation, 
and their distribution. 

Although the study was limited chiefly to native stands and full-grown 



11 For details of the economic utilization, concentrations, determinations, and yield of 
yucca, see typwritten reports filed in the Division of Cotton and Other Fiber Crops 
and Diseases, Plant Industry Station, Beltsville, Md.: 

Webber, J. M. yucca stands and districts, and estimated yields, with miscel- 
laneous notes on harvesting, fiber and costs. 76 pp., illus. 1943. Revised 1944. 

cattle and rodents in relation to economic utilization of yuccas. 

10 pp., illus. 1944. 

economic uses of yuccas in southwestern united states, with foot- 
notes on cultivation and growth rates. 25 pp., illus. 1945. 



YUCCAS OF THE SOUTHWEST 15 

nursery plants, it also included the collection of numerous specimens and their 
comparison in the laboratory. These specimens were mainly standard her- 
barium sheets of dried plants or parts of plants, dried fruit, and preserved 
flower parts. 12 

The studies of native stands, roughly speaking, extended from west-central 
Texas (from Crystal City north to Childress) to the Pacific Ocean; and from 
the Mexican border north to south-central Colorado, Utah, Nevada, and Cali- 
fornia. Within this area, numerous plants were studied throughout many parts 
of the ranges of the various species. In addition, experimental plots and work- 
ing field stations were established in several localities. 

The majority of the nursery plants were grown from seeds in open field 
plots at the California Citrus Experiment Station and the United States Re- 
gional Salinity Laboratory, Riverside, Calif. The nursery plots were subjected 
to semicultivation and to one light furrow irrigation every 8 weeks during the 
summer dry season. 

METHOD OF PRESENTATION 

In evaluating yuccas, the writer has followed the rule that species must be 
separated on the basis of more or less evident distinctions and that there must 
be little intergradation from one species to another. This does not mean that 
all species have the same degree of distinctness, as some are well denned by 
several stable characters and others are so variable that they are difficult to 
characterize. In the case of a species with variable characters it has been pos- 
sible to describe only the range of characteristics of the whole group, or, in a 
few instances, to divide them into lesser groups, varieties, or forms that exhibit 
some degree of distinctness. A species and its varieties, however, are apparently 
very similar in genetic constitution, and not uncommonly the variations found 
within the species are quite similar to the variety. 

Of the 28 to 32 species listed as occurring within the range of the present 
survey, only 21 species are recognized. This is neither extreme consolidation 
nor liberalism. It is mainly the result of the writer's species concept, which 
eliminates apparent hybrids and gradients from species status and retains 
taxonomic units clearly marked by evident characters. The most extreme con- 
solidation would be a reduction to 14 species, mainly by calling the following 
species subspecies or varieties: (1) Yucca baileyi, Y. angustisshna, and Y. 
constricta under Y. glauca; (2) Y. reverchoni under Y. rupicola; (3) Y . 
thompsoniana under Y. rostrata; (4) Y. faxoniana under Y. carnerosana; and 
(5) Y. torreyi under Y. treculeana. Such combining, however, would only 
morphologically equalize species rank. It would not reduce the number oi 
distinct taxonomic units that should bear names. 

Descriptions are based upon native plants in various sections of the species 
range. Apparent hybrids and extreme variants that are not in line with the 
general characterization of the species are not included in the descriptions. 
Likewise, plants exhibiting local environmental effects, such as roadside plants 
and plants receiving extra cattle or sheep manure, have been excluded from the 
descriptions. However, many of the apparent hybrids and variants are included 
in the general treatment of the species or species group. 

Unless otherwise noted, all measurements are as follows. Clumps: Diameter. 



12 These specimens are at the Plant Industry Station, Beltsville, Md., and the University 
of California Herbarium, Berkeley, Calif. 



16 AGR. MONOGRAPH 17, U. S. DEPT. OF AGRICULTURE 

actual measurement or careful estimate made with the aid of a meter rod ad- 
jacent to the plant; number of leaf heads (crowns), actual count or careful 
estimate based on division of clump into equal parts. Stems: Diameter meas- 
ured just above basal swelling, or approximately 30 cm. above soil level ; height, 
from ground level to center of leaf head or base of imbricated bud. Leaves: 
Dimensions of blade only; width, the greatest width, flattened width in case 
of convex leaves. Flowei' parts: Range mainly the average measurement of 5 
flowers at anthesis from a single inflorescence of each of 50 widely distributed 
plants; average (given in parentheses), the mean of the preceding range 
averages. Fruits: All measurements and descriptions are of dried matured 
fruits. Range and average measurements in general are based mainly on 50 to 
1,500 individual plants. Bajada, means an alluvial fan; upper or lower bajada, 
i.e., the upper or more elevated part of the bajada and the lower or less elevated 
part (Benson and Darrow (3) and others). 

SECTIONS OF THE GENUS YUCCA 

Fruit indehiscent. 

Leaf blade large, broad, mainly thick, concavo-convex, rigid and swordlike, or 
in Y. schottii somewhat thin and flexible; fruit large and fleshy, eventually 
becoming pendent and drying with a rough leathery, dark brown or black, 
often gummy, covering; seed rough, more or less thickened, flat, without 
marginal wing (species Nos. 1 to 8) I. SARCOCARPA. 

Leaf blade small, thin, narrow, plano-convex, rigid and stilettolike; tip concavo- 
convex; fruit dry and spongy, erect, spreading or variously directed; seed 

smooth, thin, without marginal wing (species No. 9) II. CLISTOCARPA. 

Fruit dehiscent. Leaf blade thin, mainly narrow, plano-convex, flexible, or in Y. 
rupicola and Y. gilbertiana broad and in Y. neomexicana and Y. gilbertiana, 
concavo-convex and rather rigid; fruit a dry capsule, soon becoming erect; 
seed smooth, thin flat, with or without marginal wing. 

Stigma capitate; capsule loculicidally dehiscent (species No. 10) 

III. HESPEROYUCCA. 

Stigma lobed; capsule commonly septicidally, or occasionally septicidally and 

loculicidally dehiscent (species Nos. 11 to 21) IV. CHAENOCARPA. 

KEY TO YUCCA SPECIES 

INDEHISCENT-FRUITED SPECIES 

la. Fruit fleshy; leaf blade large, coarse, and swordlike; leaf margin entire or with 

free fibers ; perianth segments thin, expanding equally or nearly in two series, 

seed thick. 

lb. Perianth segments united into slender tube; filaments inserted on tube and 

free from each other; heads of leaves symmetrical; leaf blade very rigid, 

broad, strongly concavo-convex. 

lc. Peduncle long, exceeding the foliage; panicle ellipsoidal or oblately 

ellipsoidal; perianth segments commonly united for 2.5 cm...l. Y. carnerosana. 
2c. Peduncle short, not exceeding the foliage; panicle broadly conoid; 

perianth segments rarely united for more than 1.5 cm 2. Y. faxoniana. 

2b. Perianth segments not united into tube, but slightly connate at base; 

filaments inserted at base of segments and united to each other; heads of 

leaves rather symmetrical; leaf blade commonly rigid, rarely flexible, fairly 

broad or narrow, moderately concavo-convex, or rarely somewhat* flattened. 

lc. Pistil short, 1.8 to 4.0 cm.; plant commonly fruticose or arborescent, with 

few stems and with heads of leaves in a rather open clump. 

id. Leaf margin without free fibers, or in age with a few fine straight 

fibers; ovary rather slender. 



YUCCAS OF THE SOUTHWEST 17 

le. Panicle with few branches and commonly puberulent; leaf blade 

thin and flexible 3. Y. schottii. 

2e. Panicle with many branches and commonly glabrous; leaf blade thick 

and rigid 4. y. treculeana. 

2d. Leaf margin with free fibers; fibers rather coarse and curly; ovary 
rather stout. 
le. Ovary gradually tapered from base to style; style 4 to 8 mm. long. 

5. Y. TORRE YI. 

2e. Ovary abruptly tapered at apex to style; style 1 to 2 mm. long. 

6. Y. SCHIDIGERA. 

2c. Pistil long, 4.5 to 9.0 cm.; plant acaulescent or caulescent, with numerous 
stems or with heads of leaves in rather dense clump. 
Id. Plant caulescent with rather tall, erect or assurgent stems; leaf blade 

narrow (1.8 cm.); leaf margin with few fine fibers 7. Y. arizonica. 

2d. Plant acaulescent or with short procumbent stems; leaf blade broad 

(38 cm.) ; leaf margin with coarse, often curly, fibers 8. Y. baccata. 

2a. Fruit spongy and dry; leaf blade small, fine and stilettolike; leaf margin 
minutely denticulate; perianth segments thick, fleshy, the outer series scarcely 
expanding, the inner series expanded about half their length, seed thin. 

9. Y. brevifolia. 

DEHISCENT-FRUITED SPECIES 

la. Leaf margin yellow, brown, or greenish yellow; corneous; completely or 
partially denticulate, or rarely smooth. 
lb. Stigma capitate; capsule obovoid or nearly cylindrical, tipped with an abrupt, 

short point, loculicidally dehiscent 10. Y. whipplei. 

2b. Stigma 3-lobed, 6-notched; capsule ellipsoidal or somewhat ovoid, tapering 
at apex into a long, slender point, 
lc. Plant aborescent. 

Id. Plant tall (2.3 m.) ; branches short; head of leaves large, symmetrical. 

11. Y. rostrata. 
2d. Plant medium tall (1.4 m.) ; branches comparatively long; head of 

leaves small, rarely symmetrical 12. Y. thompsoniana. 

2c. Plant acaulescent. 

id. Plant an open clump; leaf blade 20 to 25 mm. wide, undulate, dark 

green 13. Y. rupicola. 

2d. Plant a dense clump; leaf blade 10 to 20 mm. wide, not undulate, 

light glaucous green 14. Y. reverchoni. 

2a. Leaf margin at first entire, thin, paperlike, white or rarely green, but soon 

finely filiferous. 

lb. Leaf blade concavo-convex, short (15 to 30 cm.), commonly lanceolate; 

capsule small and deeply constricted, attentuate-beaked, with valve points 

outward at maturity. 

lc. Leaf blade linear to lanceolate; peduncle long, usually exceeding the 

foliage; basic flower color white 15. Y. neomexjcana. 

2c. Leaf blade lanceolate to broadly lanceolate, peduncle short, rarely exceed- 
ing the foliage; basic flower color yellow 16. Y. gilbertiana. 

2b. Leaf blade plano-convex, commonly linear, rarely lanceolate, capsule usually 
large, constricted or not constricted, mucronate-beaked, with valve points 
mainly inward at maturity, 
lc. Acaulescent or rarely with thin, short stems up to 1 m. tall; head of 
leaves commonly small. 
Id. Scape short (0.2 to 1.6 m.) ; peduncle short (0 to 1.0 m.) ; inflorescence 
a simple raceme or with a few abortive or short branchlets. 
le. Capsule rarely, and then only slightly, constricted. 

if. Style dark green, very tumid 17. Y. glauca. 

2f. Style white to fairly dark green, narrow to somewhat swollen. 

18. Y. BAILEYI. 

2e. Capsule usually constricted, constriction deep 19. Y. angustissima. 

2d. Scape long (1.5 to 3.2 m) ; peduncle long (0.7 to 2.0 m.) ; inflores- 
cence paniculate, capsules usually constricted 20. Y. constricta. 

2c. Arborescent, with thick, long stems, head of leaves commonly large. 

21. Y. elata. 



18 AGR. MONOGRAPH 17, U. S. D£PT. OF AGRICULTURE 

INDEHISCENT FRUITED YUCCAS 

1. Yucca earner osana 

(Palm barreta, palm samandoca) 

Yucca carnerosana (Trel.) McKelvey, Yuccas Southwest. U. S. 1: 24, pis. 6 
and 7. 1938. 

Yucca australis Trel., Mo. Bot. Gard. Ann. Rpt. 4: 190. 1893, as to 

Pringle specimens only. Not Y. australis (Engelm.) Trel., 1902. 
Samuela carnerosana Trel., Mo. Bot. Gard. Ann. Rpt. 13: 118. 1902. 

Plant arborescent, symmetrical, simple, or rarely a tall, rather dense clump 
of stems and heads of leaves, with stems contiguous at base and spreading 
toward top; stems commonly 1 or 2, of equal or unequal height, rarely up to 
8 of varying height, the older trunklike, up to 5 m. tall and 35 cm. in 
diameter, rarely once- or twice-branched toward top, the branches short, 
assurgent; leaves large in terminal head, the blade 51 to 100 (68) cm. long, 
5.2 to 7.5 (5.8) cm. wide, rigid, spreading; scape rather long, stout; panicle 
slightly above foliage, ellipsoidal or oblately ellipsoidal, rather densely branched 
with persistent white bracts; flowers expanding to 45 to 90 mm., white; 
sepals 67 to 94 (83) mm. long, 13 to 21 (17) mm. wide; petals 65 to 93 
(82) mm. long, 20 to 28 (26) mm. wide; tube portion of sepals and petals 
17 to 30 (25) mm. long, filaments 22 to 29 (26) mm. long; pistil 48 to 
63 (57) mm. long; ovary 6 to 9 (8) mm. in diameter; style 6 to 10 (8) mm. 
long; fruit (immature) oblong. (Pis. 1 and 2.) 

Type: Limestone hills, Carneros Pass, Carneros, northeastern Mexico, C. 
G. Pringle 3912 , May 11, 1891; Mo. Bot. Gard. Nos. 135656-57. 

Range: Only in south-central Brewster County, eastern Rio Grande Big 
Bend, Tex. Although it occasionally occurs in small patches along the lower 
bajada washes, it is usually thinly scattered on the eastern and southeastern 
foothills of the Carmen Mountains between 1,600 and 2,500 feet elevation. 
The greater part of the range of Yucca carnerosana is reported in Mexico, 
where it extends some distance below the Tropic of Cancer. 

The heads of leaves of the younger shoots are extremely large and sym- 
metrical and hemispherical in shape. The leaves are large, dark blue green, 
with very conspicuous fibers along margins of the young ones. The flower 
bud, as well as the matured inflorescence, is extremely large and showy. The 
flowers are waxen white and have a very pleasant fragrance. The robust 
growth and large inflorescence of Y. carnerosana is very attractive and can be 
seen a mile or more away. Y. carnerosana usually flowers between March 15 
and April 15, but late off-season summer and fall blooming is common. 

The species exhibits little variation. It is associated with Y. torreyi and Y. 
rostrata. Although the writer has found no evidence that Y. carnerosana hy- 
bridizes with other yuccas, McKelvey (24) cited a specimen that she believed 
to represent a hybrid between the species and Y. torreyi. 



YUCCAS OF THE SOUTHWEST 19 

2. Yucca faxoniana 

(Palma) 

Yucca faxoniana (Trel.) Sarg., Man. Trees N. Amer., p. 121, fig. 106. 1905; 
ed. 2, p. 115, fig. 111. 1922. 

Yticca australis Trel., Mo. Bot. Gard. Ann. Rpt. ~i: 190 (as to Texas 
plants). 1893. Not Y. australis (Engelm.) Trel. 1902. 

Y. australis Coult., U. S. Natl. Herbarium Contrib. 2: 436 (in part). 
1894. 

Y. macrocarpa Sarg., Gard. and Forest 8: 305. 1895, excluding synonymy. 
Not Y. macrocarpa Engelm., 1881. 

Samuela faxoniana Trel., Mo. Bot. Gard. Ann. Rpt. 13: 117. 1902. 

Plant arborescent, symmetrical, simple, or rarely a tall open clump of stems 
and heads of leaves, with stems contiguous at base and spreading toward top, 
stems commonly 1, rarely up to 6, of varying height, the older trunklike, up to 
6.5 m. tall and 40 cm. in diameter, occasionally once or twice assurgently 
branched; leaf blade 48 to 85 (59) cm. long, 4.5 to 7.0 (6.4) cm. wide, rigid, 
spreading; scape rather short and stout; panicle 0.10 to 0.25 of its length 
within foliage, broadly conoid, rather openly branched, with white or 
occasional purple-tinged persistent bracts; flowers expanding 40 to 70 mm., 
white; sepals 55 to 87 (69) mm. long, 15 to 17 (16) mm. wide; petals 54 to 
85 (68) mm. long, 19 to 20 (19.5) mm. wide; tube portion of sepals and 
petals 10 to 18 (14) mm. long; filaments 21 to 32 (27) mm. long; pistil 45 
to 59 (51) mm. long; ovary 6 to 8 (7) mm. in diameter, narrowly ovoid; style 
5 to 7 (6) mm. long; fruit 30 to 90 mm. long, 25 to 30 mm. in diameter, 
slightly tapered above and below with long (20 mm.) curved beak; seed 5 to 
8 by 7 to 10 mm., thick and flat or occasionally hemispherical, surface rough, 
dull black, wingless. (PI. 3.) 

Type: Sierra Blanca, Tex.; flowers from cut stem, Arnold Arboretum. 

Range: Eastern part of the Rio Grande Big Bend section of Texas in the 
following counties: Southern Hudspeth and Culberson; western Jeff Davis, 
and northwestern Presidio. Although the writer has seen a few plants of the 
species in the vicinity of Tesnus, eastern Brewster County, Tex., they were 
probably planted by early settlers of Tesnus. 

The species is found on rocky and gravelly mountain slopes at approximately 
3,900 to 5,000 feet elevation. On the eroding slopes and bajadas of the 
Sierra Diablo, Eagle, and Van Horn Mountains the species occurs in rather 
dense concentrations, which cover many tens of square miles. Admixed with 
Yucca faxoniana on the bajadas are Y. tor re ft and Y . baccata. In several of 
these associations about 86 percent of the plants are Y. faxoniana and 14 per- 
cent Y. torreyi, whereas Y. baccata is so thinly scattered that it averages less 
than 1 plant per acre. These concentrations frequently have from 57 to 65 Y. 
faxoniana and from 9 to 11 Y. torreyi heads of leaves per acre. On the higher 
and more rugged mountain slopes, particularly in the Eagle Mountains, Y. 
faxoniana occurs in pure stands. Occasionally at lowest elevations, where Y . 
faxoniana approaches valley floors, it is sparsely admixed with Y . elata. 

Yucca faxoniana usually flowers between April 15 and May 15. The species 
is similar to Y. carnerosana, and if the ranges of the two species overlapped it 
would be rather difficult to separate them. However, Y. faxoniana is distin- 
guishable from Y. carnerosana by its relatively shorter, broader leaves; its 



20 AGR. MONOGRAPH 17, U. S. DEPT. OF AGRICULTURE 

lower, decidedly broad, conoid-shaped panicle; and its relatively short perianth 
tube. 

Apparent Hybrids of Yucca jaxoniana and Yucca torreyi 

Approximately 5 miles west of Allamoore, Tex., and in other places where 
Yucca jaxoniana and Y. torreyi are admixed, apparent hybrids between the 
two species are quite common. The hybrids are predominantly Y. jaxoniana 
in both vegetative and reproductive characters, but resemble Y. torreyi in one 
or more characters. Plants that appear to be Y. jaxoniana frequently have 
more or less scabrous leaves, which occasionally are tapered from base to apex 
as in Y. torreyi. Other plants similar to Y. jaxoniana resemble Y. torreyi in 
their short, rather thin, weak scape. Occasionally plants that appear to be other- 
wise typical Y. jaxoniana have rather small, globose flowers, with the segments 
united only at the base, and, rarely, typical Y. jaxoniana plants produce a long, 
cylindrical, short-beaked fruit resembling Y. torreyi. McKelvey (24) reported 
hybrids between the two species. 

3. Yucca schottii 

(Mountain or hoary yucca; locally, sword cactus) 

Yucca schottii Engelm., Acad. Sci. St. Louis Trans. (1873) 3: 46. 1878. 

Plant arborescent, rather symmetrical, simple, or fairly tall (1.5 to 2.0 m.), 
rather open clump of stems and heads of leaves, with stems contiguous at base 
and somewhat spreading toward top; stems commonly 2 or 3, rarely up to 6, 
of varying height, the older trunklike, up to 4.6 m. tall and 32 cm. in diameter, 
unbranched or with 2 or 3 assurgent branches; leaves numerous in terminal 
head, or entire stem bearing fresh leaves, the blade 40 to 90 cm. long, 2.5 to 5.5 
cm. wide, divergently spreading, thin, flexible; leaf margin thin, without fibers, 
or with few fine fibers; scape very short; panicle mainly within foliage, but 
somewhat extending above, narrowly ellipsoidal, rather densely branched, very 
tomentose or rarely glabrous; flowers small, subglobose, white; perianth seg- 
ments 20 to 35 mm. long, 8 to 15 mm. wide, broadly lanceolate; filaments 12 
to 14 mm. long, slender; pistil 19 to 24 mm. long; ovary 5 to 7 mm. in diam- 
eter, abruptly tapered at apex; style about 3 mm. long; fruit 60 to 125 mm. 
long, 25 to 38 mm. in diameter, rounded at base, tapering at apex, commonly 
asymmetrical and constricted near apex; seed 5 to 8 by 7 to 10 mm., thick and 
flat or rarely rather hemispherical, rough, dull black, wingless. (Pis. 4 and 5.) 

Type: Sierras west of Santa Cruz County, Ariz., Mo. Bot. Gard. No. 135693. 

Range: Southeastern Arizona and adjacent corner of New Mexico. Along 
the southern boundary of Arizona and New Mexico, particularly in New Mex- 
ico, the species reaches its height of development in that the plants are usually 
taller, more common, and approach fairly dense concentrations. 

Yucca schottii is strictly a mountainous and foothill plant, mainly occurring 
in oak woodlands and shady canyon slopes at 4,000 to 7,000 feet elevation. 

The species is easily distinguishable by its large, broad, nonfiliferous, flexible, 
dark blue-green leaves. Although it is associated with Y. arizonica, Y . elata, 
and several transitional forms of Y. baccata-Y. arizonica, no evidence of hy- 
bridization with these plants has been found. Y. schottii flowers several months 
later than any other bacciferous yucca (July 15 to August 15), usually follow- 
ing the flowering season of Y. elata. 



YUCCAS OF THE SOUTHWEST 21 

4. Yucca treculeana 

(Palma-pita or de datils; locally, Spanish- bayonet or Spanish-dagger) 

Yucca treculeana Carr., Rev. Hort. 1858: 580. 1858. 

Yucca canaliculata Hook., Bot. Mag. 86: pi. 3201. I860. 
Y. longiftora Buckl., Phila. Acad. Sci. Proc. 1862: 8. 1863. 
Y. argospatba Verlot, Rev. Hort. 1868: 393. 1868. 
Y. aspera Engelm., Acad. Sci. St. Louis Trans. (1873) 3: 37. 1878. 
Y. treculeana var. canaliculata (Hook.) Trel., Mo. Bot. Gard. Ann. Rpt. 
13: 97. 1902. 

Plant arborescent, with 1 to few stems of varying height in rather open 
clump; stems commonly 1 to 4 assurgently branched, 3.0 to 3.5 m. tall; leaves 
commonly in large, symmetrical head, or occasionally entire shoot bearing 
fresh leaves, the leaf blade 50 to 98 cm. long, 4 to 8 cm. wide, thick, rigid, 
concavo-convex, but frequently flattened ; leaf margin thin, entire or with few 
line, straight fibers; scape short, glabrous; panicle extending 0.5 to 0.75 per- 
cent of its length above foliage, ellipsoidal, dense; flowers small, broadly 
globose or hemispherical, white or lightly tinged with purple, perianth seg- 
ments somewhat thickened toward base; sepals 29 to 45 (36) mm. long; 8 
to 15 (11) mm. wide; petals 30 to 40 (36.5) mm. long, 11 to 21 (14.5) 
mm. wide; filaments 15 to 24 (18.8) mm. long; pistil 20 to 30 (27.3) mm. 
long; ovary 4 to 6 (5.8) mm. in diameter, apex abruptly terminating 
(shouldered) in style; style 1 to 5 (3) mm. long; fruit 65 to 100 mm. long, 
17 to 24 mm. in diameter, cylindrical with abruptly tapered apex, symmetrical, 
rarely constricted; carpel-suture U-shaped, deep or frequently broad and rather 
flat; seed n 4 to 5 by 5 to 6 mm., flat, thick, rough, dull black, wingless. 
(PL 6.) 

Type: South-central Texas and adjacent Mexico; A. A. Trecul 1496, Mus. 
d'Hist. Nat., Paris. 

Range : The greater part of the reported distribution of Yucca treculeana is 
east of the present range of study and in Mexico. Consequently the writer has 
seen the species only in a limited triangular area — between Uvalde, Carrizo 
Springs, and Eagle Pass, Tex. In this triangle the species is thinly scattered in 
tall chaparral and is easily overlooked. It flowers between mid-March and early 
April. 

Although Y. treculeana is a well-characterized species, it is similar to Y. 
torreyi, and variations that occur in the Y. torreyi so closely resemble Y. trec- 
uleana that it is questionable whether specific distinctions exist between the two 
plants. Y. treculeana, however, is distinguishable from typical Y. torreyi by 
its small flowers of hemispherical shape with stout ovaries and by its more 
symmetrical head of relatively broader, shorter leaves, which are mainly non- 
filiferous. 

Apparent Hybrids of Yucca treculeana and Yucca torreyi 

Between Yucca treculeana on the east and Y. torreyi on the west is a com- 
paratively large area that appears to be a transitional zone between the two 
species. This zone is approximately bounded by a line through Eagle Pass, 



'-Seed from single plant near Rockport, Tex. 



22 AGR. MONOGRAPH 17, U. S. DEPT. OF AGRICULTURE 

Devils River (west of Del Rio) , Rocksprings, and Uvalde, Tex. Although the 
plants in this zone are quite variable, their general aspect is that of Y. torreyi. 
They commonly resemble Y. torreyi in their long, narrow leaves, slender ova- 
ries, and oblong tapered fruit; and they frequently resemble Y. treculeana in 
their nonfiliferous leaves and small, globose flowers. In this area McKelvey 
(24, p. 115) commonly found floral abnormalities, in addition to apparent 
hybrid characters, and concluded, "Some of these confusing plants may be of 
hybrid origin but, since the dominating influence in each cannot be estimated, 
all have been cited under Y. torreyi . . ." The writer has not found excessive 
vegetative or floral abnormalities in the region, and he is of the opinion that 
the plants are not so variable as McKelvey indicates. The two species are quite 
similar, and the variability that occurs is within the range to be expected of 
hybrids and hybrid recombinations. 

5. Yucca torreyi 

(Torrey yucca) 
Leaf blade straight, rigid, glabrous or glaucescent, yellow green in color 

Y. TORREYI. 

Leaf blade falcate, rather flexible, glaucous, blue green in color.. ..Y. torreyi, blue form. 
Yucca torreyi Shafer, in Britton and Shafer, N. Amer. Trees, p. 157, fig. 117. 
1908. 

Yucca baccata var. macrocarpa Torr., in Emory, U. S. and Mex. Bound. 

Bot., p. 221. 1859. 
Y. macrocarpa (Torr.) Merriam, N. Amer. Fauna 7: 358 (name only). 

1893; Cov., U. S. Natl. Herbarium Contrib. 4: 202. 1893; Trel., Mo. 

Bot. Gard. Ann. Rpt. 13: 110. 1902. Not Y. macrocarpa in Engelm., 

1881. 
Y . torreyi f. parpi flora McKelvey, Yuccas Southwest. U. S. 1: 112. 1938. 

Plant caulescent, rather commonly arborescent, asymmetrical, or of ragged 
appearance, simple to rather tall, narrow, fairly dense clump of stems and 
heads of leaves; stems commonly 1 or 2, occasionally up to 8, of varying height, 
the older trunklike, rarely up to 4.3 m. tall and 30 cm. in diameter, unbranched, 
or rarely with 2 or 3 ascending branches; leaf head solitary or somewhat clus- 
tered; leaves few in small, terminal head or numerous in large, elongated head 
or cluster, the blade 30 to 103 (58.7) cm. long, 2.8 to 5.0 (3.6) cm. wide, 
commonly tapered from enlarged base to apex, moderately concavo-convex, 
or rarely somewhat flattened, thick, rigid, scabrous on both surfaces, yellowish 
green; leaf margin thick, with at first curly, later straight, tough fibers; scape 
to 10 (5.2) cm. long; panicle commonly 0.1 to 0.5 of its length above 
foliage, or rarely entirely within, 36 to 70 (46) cm. long, ellipsoidal, dense; 
flowers commonly subglobose or campanulate, occasionally fully expanding, 
cream color to deeply tinged with dark purple; perianth segments very variable 
in shape and size, concave, thickened in the center from base to apex; sepals 
34 to 75 (52) mm. long, 8 to 18 (144) mm. wide; petals 33 to 78 (50.9) 
mm. long, 10 to 22 (17.8) mm. wide; filaments 14 to 27 (18.5) mm. long; 
pistil 22 to 35 (28.2) mm. long; ovary 4 to 8 (5.9) mm. in diameter, rather 
stout, apex tapering into style; style 4 to 8 (5.8) mm. long; fruit 70 to 105 
mm. long, 25 to 38 mm. in diameter, cylindrical or ovoid, gradually tapered 
from base to apex, rarely constricted ; carpel suture V-shaped, deep, and narrow; 
seed 5 to 8 by 6 to 9 mm., flat, thick, rough, dull black, wingless. (PI. 7.) 



YUCCAS OF THE SOUTHWEST 23 

Type: Locality near Limpia Creek and near Presidio del Norte, Tex., col- 
lected by John Bigelow, the specimens apparently not now extant. 

Range: Sierra and Dona Ana Counties, N. Mex., to the Devils River, Val 
Verde County, Tex., and south; but it is found as far north as Carlsbad, N. 
Mex. The species is usually thinly scattered on grassy and chaparral mesas and 
slopes at 2,000 to 5,000 feet elevation. It has not been found in dense con- 
centration on any large area. Possibly the best and most typical plants ot 
Yucca torreyi occur in Jeff Davis County and in the vicinity of Presidio, Tex. 
It usually flowers between late March and early May. 

The species is rather variable throughout its range and blends with, or 
exhibits characters transitional to, Y. treculeana on its eastern boundary 
(Maverick, Kinney, Uvalde, and Edwards Counties, Tex.) and with Y. 
baccata on its western boundary (Grant, Sierra, and Dona Ana Counties, N 
Mex.). Probably the greatest variation in the species is in the size and shape 
of its flowers and in the height, size, and density of its panicles. 

Yucca torreyi forma parviflora McKelvey is mainly characterized by its less 
branched panicle and its smaller, globose flowers. Since the species and form 
can be distinguished only by separating the extremes of the characters involved, 
the form is not worthy of a formal name. 

Blue Form of Yucca torreyi 

While at Black Gap in 1944, V. L. Cory 14 called the writer's attention to a 
new form of Yucca torreyi. In general, this form exhibits the distinctions that 
commonly characterize blue forms of fleshy-fruited yuccas. These blue form 
characters are as follows: 

Plants mainly a denser, more confused cluster; stems more numerous and 
shorter; heads of leaves more asymmetrical; leaves commonly longer, always 
narrower, usually twisted and variously curved, always a sage or blue-green 
color and usually quite glaucous. 

The form apparently extends north and east from Black Gap to the Pecos 
River basin. 

In addition to typical blue-form characters, the form has smaller flowers 
than the species and they are usually rather heavily tinged with purple. The 
latter characters, however, are not necessarily distinguishing characters of the 
form, as fairly typical Y. torreyi with very small, extremely purple-tinged 
flowers occurs only a few miles west of Black Gap near Persimmon Gap in the 
Santiago Mountains. 

Apparent Hybrids of Yucca torreyi and Yucca baccata 

Yucca torreyi is rather easily distinguished from closely related species by 
its relatively ragged appearance and its comparatively long, narrow, yellowish- 
green leaves. However, in localities where plants of Y. torreyi and Y. baccata 
are rather thickly intermingled, the species are difficult to separate and ap- 
parent hybrids are quite common. In these areas tall plants with one or two 
vertical stems and acaulescent, fairly dense, clumped plants occur side by side. 
Both the tall and the clumped plants frequently have rather short, broad, dark- 
green leaves, and aside from height and clump habit, the majority of them 



14 See footnote 4, p. 5. 



24 AGR. MONOGRAPH 17, U. S. DEPT. OF AGRICULTURE 

are vegetatively inseparable. Upon flowering, the tall-stemmed plants always 
exhibit panicles and flowers that resemble Y. torreyi, whereas the clumped 
plants exhibit either typical Y. baccata panicles and flowers or a panicle of 
baccata height, with an intermediate number of panicle branchlets. The flowers 
of the apparent hybrids are considerably larger than the average Y. torreyi 
flower, but their pistils are midway in length between the two species. 

Although McKelvey (24) found no evidence of a crossing between Y. 
torreyi and Y. baccata, she observed characters that suggest a remote relation- 
ship between the species. Benson and Darrow (3) placed Y. torreyi in Mc- 
Kelvey's Baccatae series, as Y. baccata var. macrocarpa Torr., but they noted 
that V. L. Cory and McKelvey believed it to be a distinct species, more clearly 
related to Y. treculeana. Although the writer also holds the latter opinion, he 
believes that, except for Y. schottii, there is little difference between McKel- 
vey's Baccatae and Treculeanae series. 

Apparent hybrids involving V. torreyi and other species have been dis- 
cussed under Y, carnerosana, Y. jaxoniana, and Y . treculeana. 

Although Y. torreyi is commonly admixed with Y. elata and occasionally 
with Y. thompsoniana and Y. reverchoni, no evidence of hybridization with 
these species has been observed. 

6. Yucca schidigera 

(Mohave yucca) 

Leaf blade broad, straight, very rigid, and yellow green in color Y. schidigera. 

Leaf blade narrow, falcate, rather flexible, and blue green in color 

Y. schidigera, blue form. 

Yucca schidigera Roezl ex Ortgies, Gartenflora 20: 110. 1871. 

Yucca calif omica Nutt. ex Baker, Linn. Soc. London, Jour., Bot. 18: 229 

(as syn. Y. baccata Torr.). 1880. 
Y. macrocarpa Merriam, N. Amer. Fauna 7: 358. 1893; Cov., U. S. 

Natl. Herbarium Contrib. 4: 202 (Calif, and Nev. plants). 1893. 

Not Y. macrocarpa Engelm. 1881. 
Y. mohavensis Sarg., Gard. and Forest 9: 104. 1896. 

Plant commonly fruticose, with rather symmetrical appearance, or clumped 
with ragged appearance; clump rather tall, broad and open, with at least several 
stems of equal height; stems occasionally 1 to 3, commonly 4 to 7, rarely up 
to 23, erect or somewhat assurgent, the older trunklike, rarely up to 2.5 m. 
tall and 40 cm. in diameter, simple or 1 to' 4 ascending erect branches; head 
of leaves usually solitary, rarely clustered; leaves numerous, the blade 33 to 105 
(56) cm. long, 2.5 to 5.0 cm. wide, broadest near middle, the greater part 
rather deeply concavo-convex, thick, very rigid, yellow green; leaf margin 
thick with coarse, somewhat curled fibers; scape to 15 cm. long; panicle 
entirely within foliage, or to 0.5 of its length above, 50 to 125 cm. long, ellip- 
soidal or with apex flattened, densely branched and flowered; flowers globose, 
white or cream, commonly tinged with lavender or purple; perianth segments 
lanceolate or broadly lanceolate; 24 to 45 mm. long, 6 to 10 mm. wide; fila- 
ments 13 to 16 mm. long; pistil 18 to 25 mm. long; ovary 5 to 8 mm. in di- 
ameter, rather stout, tapering into style; style 1 to 2 mm. long; fruit variable, 
long and cylindrical, 90 to 115 mm. long, 30 to 38 mm. in diameter, with 75 
percent medium-constricted, or short variable shape, but mainly tapering from 
swollen base to rather blunt apex 60 to 85 mm. long, 25 to 35 mm. in di- 



YUCCAS OF THE SOUTHWEST 25 

ameter; seed 6 to 9 by 8 to 11 mm., flat, thick, rough, dull black, wingless. 
(Pis. 8 and 9.) 

Type: San Diego, Calif. Nut tall, Gray Herbarium; type of Yucca calif or - 
nica Nutt. 

Range: Western half of the Californif-Baja California border, north 
through San Diego, Riverside, and San Bernardino Counties, Calif., thence 
east and northeast, crossing the California boundary into the northwest quarter 
of Mohave County, Ariz., and into southern Clark County, Nev. Y. schidigera 
occurs on gravelly mountain and valley slopes of desert and chaparral, at 
1,000 to 6,000 feet elevation. Throughout most of its southern California 
range, Y. schidigera is thinly scattered and rarely forms dense concentrations 
of large area. On the other hand, in northwestern and west-central 
Mojave Desert in California, in Nevada, and in the Sacramento and northern 
Hualpai Valleys, Ariz., it occurs in dense concentrations, wMch cover many 
tens of square miles. These stands frequently have 111 to 286 (144.12) heads 
of leaves per acre (pi. 8). In a letter to the writer, dated April 14, 1947, Nell 
Murbarger, of Costa Mesa, Calif., reported that the species occurred at 7,800 
feet, above Baldwin Lake in the San Bernadino Mountains on the slope from 
Gold Mountain toward Lucerne Valley. The species usually flowers between 
April 1 and May 15. 

In many parts of California Y. schidigera is associated with Y. whipplei, 
and, to a lesser degree, with Y. baccata var. vespertina and Y. brevifolia. In 
Arizona and Nevada it is commonly intermingled with Y. baccata var. vesper- 
tina and Y. brevifolia. 

Blue Form of Yucca schidigera 

Yucca schidigera is well denned and throughout most of its range exhibits 
little variation. A typical blue form (p. 23) centers in the Morongo Valley, 
San Bernardino County, Calif., and spreads eastward through Yucca Valley 
toward Twentynine Palms and westward through San Gorgonio Pass to the 
western part of the Salton Sea basin, Riverside County, Calif. 

Apparent Hybrids of Yucca schidigera and 
Yucca baccata variety vespertina 

In localities where Yucca schidigera and Y. baccata var. vespertina inter- 
mingle, unquestionable hybrids, and especially plants of hybrid derivation, 
are common. Such confusing plants are particularly abundant on the western 
bajada of the Ivanpah Mountains, San Bernardino County, Calif., and on the 
eastern bajada of the White Hills, Mohave County, Ariz. The majority of the 
characters of the hybrids are midway between the two species, although the 
characters of the more confusing plants vary considerably in their likeness to 
Y. schidigera or Y. baccata var. vespertina. 

Comparison of Yucca schidigera, Yucca torreyi, 
Yucca treculeana, and Yucca schottii 

Relative to Yucca schidigera, Y. torreyi, Y. treculeana, and Y. schottii 
McKelvey (24, p. 66) stated, "No distinctions in habit of the four plants 
holds good universally; ... in the writer's mind, however, the impression 



26 AGR. MONOGRAPH 17, U. S. DEPT. OF AGRICULTURE 

persists that plants of Y. Treculeana and Y. Schottii zre of a more pleasing, 
finer (that is less coarse) appearance than plants of Y. Torreyi and Y. scbidi- 
gera; . . . Yucca scbidigera and Y. Schotth commonly produce several stems 
and, in the last, one stem is apt to incline outward from the base; Y. Torreyi 
and Y. Treculeana as a rule are single-stemmed and average taller than the 
other members of the group; when the leaves have fallen from the trunk of 
Y. Treculeana it looks over slender for the size of the plant." In general, this 
writer concurs with McKelvey, but he does not agree that Y. schottii produces 
more or shorter stems than Y. torreyi. In practically all yuccas, variation in 
height and in number of stems is quite common. Not only do variations occur 
in one locality, but different localities may exhibit equal, less, or greater ranges 
of variation, with different means. Thus, Y. torreyi of western Rio Grande 
Big Bend, Tex., would unquestionably average several feet taller and several 
stems less per plant than those of eastern Big Bend, Tex. ; whereas Y . schottii 
in the Peloncillo Mountains and Animas Valley, N. Mex., are taller and have 
fewer stems than those plants of the species along the boundary of Pima- 
Cochise Counties, Ariz. 

Yucca scbidigera has more and shorter stems than Y. torreyi. and it is usually 
more fruticose or shrubby than Y. torreyi, Y. schottii. and Y. treculeana. 
When, rarely, Y. scbidigera is arborescent, it is dwarf, compact, and massive, 
with shorter, more ascending branches than the comparable species. 

7. Yucca arizonica 

Yucca arizonica McKelvey, Arnold Arboretum Jour. 16: 270. 1935. 

Yucca puberula sensu Torrey, in Emory, U. S. and Mex. Bound. Bot., 

p. 221. 1859, in part. Not Haworth 1828. 
Y. brevifolia Schott. ex Torr., in Emory, U. S. and Mex. Bound. Bot., p. 

221. 1859, in part, as synonym of Y. puberula. Not Y. brevifolia 

Engelm. 1871. 
Y. treleasei Macbride, Gray Herbarium, Harvard Univ., Contrib. 56: 15. 

1918. Not Y. treleasei Sprenger. 1901. 
y. baccata var. brevifolia Benson & Darrow, Amer. Jour. Bot. 30: 234. 

1943, in part. 

Plant caulescent, forming tall, rather broad, ragged, fairly dense clump; 
clump 1.2 to 3.7 m. in diameter, with 6 to 43 (19.95.) heads of leaves at a 
density of 1.2 to 3.0 (1.9) leaf heads per square meter; stems, the oldest 
mainly assurgent or occasionally a few erect or rarely procumbent, of varying 
height, up to 1.0 to 2.4 (1.6) m. tall, occasionally 1 to 2 short-branched; leaf 
heads commonly solitary, rarely clustered; leaves rather numerous, the blade 
30 to 65 (43.9) cm. long, 1.4 to 3.0 (1.9) cm. wide, slightly broadened 
toward middle, straight or somewhat incurved, mainly concavo-convex, some- 
what flexible; leaf margin thin with fine, straight fibers; scape 10 to 32 (20.7) 
cm. long; panicle entirely above foliage to 0.6 (0.8) of its length above, long 
and narrow, 43 to 68 (55.6) cm. long, narrowly obovoid; flowers campanulate 
and not expanding, or very rarely somewhat subglobose and slightly expand- 
ing, white or cream, commonly tinged with purple; perianth segments lanceolate 
to rarely oblanceolate; sepals 55 to 122 (75.9) mm. long, 13 to 28 (21.9) mm. 
wide; petals 60 to 124 (75.3) mm. long, 22 to 36 (29.7) mm. wide; filaments 
36 to 59 (43.3) mm. long; pistil 46 to 74 (57.7) mm. long; ovary 7 to 10 
(8) mm. in diameter, narrowly ovoid, broad at base and tapering into style, 



YUCCAS OF THh SOUTHWEST 27 

with deep carpel sutures and commonly faint anther depressions; style 4 to 10 
(5.5) mm. long, tapered; fruit 70 to 120 mm. long, 20 to 30 mm. in diameter, 
symmetrical, broadly cylindrical or somewhat tapering; seed fairly large, flat, 
thick, rough, dull black, wingless. (PI. 10.) 

Type: Probably near Nogales, Ariz.; Mo. Bot. Gard. No. 135693. 

Range: Southeastern half of Pima County, Ariz., east through Santa Cruz 
and Cochise Counties, Ariz., into the southwestern corner of Hidalgo County, 
N. Mex. The species occurs on gravelly and rocky hills, mountain slopes, and 
mesas of desert grassland and oak woodland at 3,000 to 5,700 feet elevation. 
In the western limits of its range, particularly on the slopes of the Baboquivari 
Mountains, the species occurs in fairly dense concentrations that cover many 
square miles. Eastward it becomes less abundant, until in the vicinity of the 
Arizona-New Mexico line it occurs intermittently in small patches of thinly 
scattered plants. The species usually flowers between April 15 and May 15. 

Yucca arizonica is well defined, and in the southern part of its range exhibits 
little variation. Although the foliage of the species is usually somewhat blue 
green, about 5 percent of the plants in the northern part of the Baboquivari 
Mountains exhibit a more pronounced blue color and approach the typical 
yucca blue form (p. 23). 

In the south-central and southeastern part of its range, Y. arizonica is some- 
what admixed with Y. schottii, and along its northern border it is occasionally 
associated with Y. elata. In several areas north and east of the range of Y. 
arizonica, the yuccas exhibit characters that appear to be of a transitional nature 
between Y. arizonica and Y. baccata. These plants are discussed under Y . 
b ace at a. 

8. Yucca baccata 

(Datil; locally, banana yucca) 

Leaf blade broad, straight, dark green, glabrous Y. baccata. 

Leaf blade rather narrow, falcate, blue green, glaucous Y. baccata var. vespertina. 

Yucca baccata Torr., in Emory, U. S. and Mex. Bound. Bot, p. 221. 1859. 

Plant commonly simple or clumped at ground level or rarely caulescent, 
with very short assurgent or procumbent stems toward center of clump; clumps 
1 to 5 m. in diameter, with 6 to 70 (27.3) heads of leaves at density of 3-9 
to 6.8 (5.0) leaf heads per square meter; leaf blade broadened toward middle, 
commonly straight or incurved, rarely outcurved, occasionally twisted, rather 
deeply concavo-convex, quite rigid, 30 to 71 cm. long. 3.0 to 5.5 cm. wide; 
leaf margin usually with coarse, short recurved fibers, occasionally with long 
curly fibers; scape to 13 (7.0) cm. long; panicle entirely within foliage to 
0.5 of its length above, obovoid or quite ellipsoidal, with oblate or acute apex, 
36 to 60 (46.9) cm. long; flowers campanulate, expanding but little, pendent, 
white or cream, commonly tinged with purple; perianth segments lanceolate 
to oblanceolate; sepals 40 to 99 (66.3) mm. long, 14 to 28 (18.7) mm. wide; 
petals 43 to 95 (66.9) mm. long, 18 to 30 (22.6) mm. wide; filaments 24 to 
40 (31.4) mm. long; pistil 38 to 74 (48.6) mm. long; ovary narrowly ovoid 
and tapering into style at apex, with deep carpel sutures, 7 to 10 (7.6) mm. 
in diameter; style 4 to 10 (6.1) mm. long, tapered; fruit 70 to 235 mm. long, 
25 to 55 mm. in diameter, symmetrical, broadly cylindrical, or somewhat taper- 
ing; frequently weighing 425 to 500 gm.; seed 6 to 9 by 7 to 11 mm., flat 
thick, rough, dull black, wingless. (PI. 11.) 



28 AGR. MONOGRAPH 17, U. S. DEFT. OF AGRICULTURE 

Type: Hurrah Creek, Guadalupe County, N. Mex., 1853, /. M. Bigelow; 
U. S. Natl. Herbarium No. 35875. 

Range: Eastern Mojave Desert, San Bernardino County, Calif., east through 
the southern tip of Nevada (Clark and Lincoln Counties) ; southwestern 
corner of Utah (Beaver, Iron, Washington, and Kane Counties) ; the greater 
part of Arizona (all except southern fourth of State) ; all of New Mexico; 
south-central and southwestern Colorado; and finally southeast through Texas 
to the vicinity of Shannon ranch, near Ozona, Crockett County, Tex. 15 Yucca 
baccata is usually scattered on hill and mountain slopes and plains of grass- 
land, pinyon- juniper, and oak woodlands at 2,500 to 7,250 feet elevation. 
Although the species usually flowers between April 15 and June 1, flowering 
is frequently delayed at higher elevations. Possibly the largest and densest 
stand of Y. baccata occurs along the western bajadas of the San Andres Moun- 
tain near Rhodes Pass, Socorro County, N. Mex., and variety vespertina is 
in scattered patches on the mesa between Kingman and Peach Springs, Ariz. 

Yucca baccata is frequently associated with Y. schidigera and Y. torreyi and 
less commonly with Y. brevifolia, Y. jaxoniana, and possibly with Y. treculeana 
and Y. arhonha. In addition, it is commonly admixed with several capsular- 
fruited yuccas. Apparent hybrids and hybrid segregates between Y. baccata 
and Y. torreyi have been cited on page 23, and those between Y. baccata and 
Y. schidigera on page 25. Apparent hybrids between Y. baccata and Y. arizonica 
are discussed on page 29. A reported hybrid between Y. baccata and Y. glauca 
(capsular-fruited species) is discussed under Y. glauca. 

Yucca baccata variety vespertina 

Yucca baccata var. vespertina McKelvey, Yuccas Southwest. U.S. 1: 45. 1938. 

Type: Just east of Peach Springs, Ariz., 1931, McKelvey 2167. 

Intermingled with Yucca baccata in the western half of its range is the 
variety vespertina, which exhibits the typical yucca blue-form characters (p. 23). 
(Pis. 12 and 13.) 

Yucca baccata is one of the most variable species of the indehiscent-fruited 
yuccas. McKelvey (24) described the habit of Y. baccata (p. 30) as, "Plant 
commonly simple, acaulescent, occasionally forming small open clumps with 2-6 
short, procumbent stems and heads of leaves. Stems scarcely exceeding 1 m. 
in length, covered with dead, reflexed leaves ..." and that of variety vesper- 
tina (p. 45) as, "Plant acaulescent with many short, procumbent or slightly 
longer, more erect stems and forming dense, confused clumps." 

It is the writer's opinion that except for the confused nature of the varietal 
clumps, the habit of both species and variety are similar and that clumps with 
a few heads of leaves are more common than solitary plants. Furthermore, it 
is believed that Y. baccata rarely forms strictly aerial stems and that their for- 
mation is limited to the centers of dense clumps. Such plants as exhibit notice- 
ably long or definite aerial stems, whether assurgent or procumbent, are found 
mainly — possibly only — in areas where definite stemmed species, such as 



15 Shannon ranch is considerably farther east than Y. baccata has previously been re- 
ported. In this vicinity, near Five Oak Draw, typical Y. baccata occurs on a mesa, at 
approximately 2,500 feet elevation. Although the writer visited only one mesa in this 
vicinity, it is very likely that Y. baccata occurs on other mesas, which are common in 
the region. V. L. Cory, who acted as the writer's guide to the mesa, refers to it as 
"Salviastrum Mesa" (9). 



YUCCAS OF THfc SOUTHWEST 29 

Y. scbidigera, Y. torreyi, or Y. arizonica, occur. In the vicinity of Kingman, 
Mohave County, Ariz., where the ranges of Y. baccata var. vespertina and Y. 
scbidigera overlap, open clumps of long-stemmed plants of Y. baccata var. 
vespertina are rather common; whereas, not 10 miles east, on the Haulpai Valley 
Mesa, the variety occurs in pure stands of dense clumped plants that are stem- 
less or practically stemless. The same condition occurs in the Mojave Desert, 
Calif., and a similar behavior has been noted in the case of Y. baccata inter- 
mingled with Y. torreyi north of Las Cruces, N. Mex. Except in Mohave 
County, Ariz., Y. baccata is the only indehiscent-fruited yucca that occurs north 
of south-central Arizona, New Mexico, and Texas, and all plants of the species 
in the considerably larger northern range are usually without aerial stems. 

Apparent Hybrids of Yucca baccata and Yucca arizonica 

Yucca thomberi McKelvey, Arnold Arboretum Jour. 16: 268. 1935. 

Yucca confinis McKelvey, Yuccas Southwest. U. S. 1: 49. 1938. 
Y. baccata var. brevi folia Benson & Darrow, Amer. Jour. Bot. 30: 234 
(in part). 1943. 

It is the writer's opinion that the large area separating Yucca baccata on the 
north and east and Y. arizonica on the south and west is a transitional zone 
between the two species. This zone occurs in the bordering quarters of Gra- 
ham, Pinal, and Pima Counties, Ariz., and extends east and southwest through 
Cochise County, Ariz., into the southwestern corner of Hidalgo County, N. 
Mex. The area is populated with scattered variable plants and scattered dense 
patches of fairly uniform yuccas. Detailed and minor studies in many locali- 
ties of the area show that the yuccas are similar to Y. baccata or Y. arizonica, 
or are midway between the two species in varying degrees. In general the 
plants in the eastern part of the area resemble Y. baccata in clump size and 
density, leaf shape, and panicle shape and size; but the stems and panicle 
height of these plants are too tall for the species. In the central part of the 
area plants resemble Y. baccata in clump size and leaf shape; Y. arizonica in 
clump density and panicle shape and size; and are about intermediate in panicle 
and stem height (pi. 14) . In the western part of the area the yuccas approach 
Y. arizonica in leaf shape and panicle height, but the clump size and density is 
more like Y. baccata and the stems are intermediate in height. Although the 
flower characters of the two species are quite similar, measurements show that 
the flowers in the eastern part of the zone resemble those of Y. baccata, and 
those in the western part are similar to Y. arizonica. 

Adding to the complexity of the plants in this zone are typical blue-form 
characters (p. 23). This character enters the zone from Safford and spreads 
northwest to approximately Miami, Ariz., where it appears to reach its height. 
The blue-form type does not occur in the southern part of the zone. It is the 
writer's opinion that failure to recognize the blue-form characters has caused 
much of the confusion as to the taxonomic position of the yuccas in north- 
eastern Pinal and north-central and northwestern Graham Counties, Ariz. 

Under Baccatae McKelvey (24) described four species. Two of these 
species, Y. baccata and Y. confinis, were distinguished (24, p. 14) by their 
acaulescent habit and by the "clavate tip of filament turning at a more or less 
conspicuous angle to the lower portion at anthesis." The other two, Y. ariz- 
onica and Y. thomberi, were characterized by their caulescent nature and by the 
"clavate tip of filament little angled in relation to lower portion at anthesis." 



30 AGR. MONOGRAPH 17, U. S. D£PT. OF AGRICULTURE 

McKelvey separated Y. con finis and Y. arizonica from Y. baccaia and- V. 
thornberi by the ligneous inflorescences of the former. According to McKelvey 
the inflorescence of Y. confinis extended about one-half of its length above 
the foliage and that of Y. arizomca extended beyond the foliage for the greater 
part of its length. The inflorescences of Y. baccata and Y. thornberi were 
fleshy at an thesis; that of Y. baccata was nearly buried in the foliage; and that 
of Y. thornberi was raised above the foliage for one-third to one-half of its 
length. 

Yucca confinis occurs in the extreme southeastern corner of Arizona and 
adjacent corner of New Mexico, and Y. thornberi occurs throughout Cochise, 
northeastern Pima, and southwestern Pinal Counties, Ariz. In these plants the 
writer has been unable to find any differences of the angle of the filament tip 
at anthesis and, although in several instances the inflorescence appears more 
ligneous than in others, in the main, woodiness is a variable character and 
more or less fleshy and ligneous inflorescences occur in ail four species. 

Aside from the presence or absence of stems, the writer has been unable 
to separate Y. confinis and Y. thornberi. In the type locality of Y. confinis, 
both stemmed and acaulescent plants occur side by side. Although scattered 
plants and patches of plants closely adhere to Y. thornberi, the writer is of the 
opinion that they, as well as the few plants adhering to Y. confinis, are transi- 
tional plants. 

According to McKelvey (24), her studies indicated that in the southeastern 
part of the range of Y. thornberi the species probably hybridized with Y. 
arizonica, Y. confinis, and Y. baccata; whereas, in its northern distribution it 
might hybridize with Y. baccaia, Y. baccata var. vespertina, and probably with 
Y. arizonica, although the range of Y. arizonica was widely separated from 
that of the northern range of Y. thornberi. Relative to this group of species 
McKelvey (24, p. 63) stated, "The flowers of all four species of the group 
f Baccatae'] are very similar but marked difference of habit, of inflorescence 
and, to a lesser degree, of foliage distinguish them." 

Benson and Darrow (3) included Y. arizonica, Y. thornberi, and Y. confinis 
in Y. baccata Torr. var. brevi folia (Schott) Benson & Darrow; whereas Kear- 
ney and Peebles (20) placed Y. confinis, along with Y. baccata var. vespertina, 
under Y. baccaia, and Y. thornberi under Y. arizonica. The distinctions be- 
tween Y. baccata and Y. arizonica are well marked at the extremities of their 
ranges, but transitional forms predominate in the area of their overlapping 
ranges. 

9. Yucca brevi folia 

(Joshua-tree; locally, tree-yucca, cactus-yucca, yucca-palm) 

Plants solitary or rarely with 2 to 3 stems. 

Stems 5 to 12 m. tall; openly branched at 2 to 3 m. above soil level 

Y. BREVIFOLIA. 

Stems 3 to 6 m. tall; densely branched at 0." to 1.0 m. above soil level 

Y. BREVIFOLIA var. JAEGERIANA. 

Plants clumped with numerous stems Y. brevifolia forma herbertii. 

Yucca brevifolia Engelm. in King, U. S. Geol. Expl. 40th Par. Rpt. 5: 496. 
1871. 

Yucca draconis var. arborescens Torr., Pacific R. R. Rpt. 4:147. 1857. 

Y. arborescens (Torr.) Trel., Mo. Bot. Gard. Ann. Rpt. 3: 163. 1892. 



YUCCAS OF THE SOUTHWEST 31 

Clistoyucca arborescens (Torr.) Trel., Mo. Bot. Gard. Ann. Rpt. 13: 41. 

1902. 
C. brevijolia (Engelm.) Rydb., Flora Rocky Mountains and Adj. Plains, 
p. 170. 1917. 

Plant arborescent, commonly with 1, occasionally 2 or 3 huge, trunklike 
stems; stems 5 to 12.5 m. tall, 0.4 to 0.8 m. in diameter, expanding rapidly 
to extremely large base, mainly dichotomously branched; primary branches 
erect-assurgent; secondary branches diffused; leaves numerous, encircling ends 
of branches, tapering from base to pungent apex, plano-convex or triquetrous, 
rigid, the blade 15 to 35 cm. long, 0.7 to 1.5 cm. wide; leaf margin thin, 
horny, minutely denticulate; scape to 8 cm. long; panicle 25 to 55 cm. long, 
ellipsoidal or ovoid, very dense; flowers oblate-ellipsoidal to globose, scarcely 
expanding, greenish white; perianth segments thick and succulent, oblong to 
lanceolate with incurved apex, the outer series scarcely expanding, the inner 
series expanding about half their length; sepals 27 to 59 (45.9) mm. long, 11 
to 20 (18.8) mm. wide; petals 26 to 65 (43.9) mm. long, 13 to 22 (14.4) 
mm. wide; pistil 20 to 34 (27.5) mm. long; ovary 8 to 11 (9.6) mm. in di- 
ameter, tapering from base to apex, pale green, with shallow carpel sutures 
and deep anther depressions; style to 5 mm. long; filaments 10 to 20 (11.6) 
mm. long, stout, stiff; fruit 60 to 85 mm. long, 30 to 45 mm. in diameter, dry 
and rather spongy, ellipsoidal, commonly with deep fissures at carpel sutures; 
seed 8 to 11 by 9 to 12 mm., flat, slightly thickened, smooth or rarely some- 
what wrinkled; dull black, and wingless. (PL 15.) 

Type: Date Creek, Yuma County, Ariz., 1869, with Parry note attached, 
Palmer, Mo. Bot. Gard. Nos. 135643, 135646. 

Range: California — from the Little San Bernardino, Cottonwood, and Iron 
Mountains, Riverside and San Bernardino Counties, northwest through the 
Mojave Desert to Owens Valley, Inyo County; Nevada — from Esmeralda 
County east and southeast through Nye, Lincoln, and Clark Counties; Utah 
— southwestern Washington County ; and Arizona — western Mohave, south- 
western Yavapai and northern Yuma Counties. 

Yucca brevijolia, its variety, and its form occur on desert plains and alluvial 
fans at 1,500 to 6,000 feet elevation. At the lower elevations it is commonly 
associated with the creosotebush, and at higher elevations with junipers, oaks, 
and pinyon pines. In a letter Nell Murbarger (p. 25) stated that on the slopes 
of Gold Mountain, San Bernardino Mountains, Y. brevijolia was not uncom- 
mon among pinyon pine and juniper at 6,500 feet elevation. Although the 
species was frequently scattered, it, like other yuccas, commonly occurred in 
definite belts and communities. Photographs supplied by Miss Murbarger 
indicate that the plants at the high elevation are forma herbertii. Y. brevijolia 
and its variety are the most imposing yuccas in the Southwest and rank with 
the giant cactus (Cereus giga??teus) as a symbol of the desert. Individual 
plants 15 to 18 m. or more tall have been reported. 

Yucca brevijolia variety jaegeriana 

Yucca brevi folia var. jaegeriana McKelvey, Arnold Arboretum Jour. 16: 269. 
1935. 
Yucca brevijolia var. woljei Jones, West. Bot. Contrib. 18: 125. 1935. 
Variety jaegeriana differs from the species in its smaller growth habit, 3 to 



32 AGR. MONOGRAPH 17, U. S. DEPT. OF AGRICULTURE 

6 m. tall; more numerous, shorter, and rarely spreading branches that develop 
closer to the ground; and shorter leaves, 10 to 20 cm. long. The dwarf, compact 
nature of the variety is unique and attractive; in many respects it gives the 
impression of an oriental dwarf tree. (Pis. 16 and 17.) 

Type: Vicinity of the Shadow Mountains, San Bernardino County, Calif., 
about 1,200 m., April 30, 1932, McKelvey 2132, Arnold Arboretum. 

Range: From Shadow, Kingston, Clark, and New York Mountains, north- 
eastern San Bernardino County, Calif., east and northeast across Clark County. 
Nev., into Washington County, Utah, and Mohave County, Ariz. 

Wolf {35, p. 7), Munz (27), and others called attention to two large 
forests of variety jaegeriana that far surpassed any other yucca concentration 
in the southwestern United States. One of these concentrations lies between 
the southern extension of the Shadow Mountains and the Ivanpah Mountains 
and extends west to Halloran Springs and east into the Ivanpah Valley about 
Cima, Calif. The other is southeast of the preceding concentration, extending 
from the western side of the New York Mountains eastward through Lanfair 
Valley into the southern end of the McCullough Mountains, Nev. (pi. 17). 

Yucca brevifolia forma berbertii 

Yucca brevifolia forma berbertii, forma nov. 16 

Forma berbertii differs from Yucca brevifolia and its variety jaegeriana in 
its extremely large, dense clump of many shoots and heads of leaves. A single 
clump frequently covers an area 8 m. in diameter, with 30 to 40 trunklike 
stems. In the majority of plants, the older stems are toward the center of the 
clump and are from 3 to 5 m. tall. Such stems bear from 4 to 12 short branches, 
and they frequently recline on adjacent stems, as if too weak to stand. (PI. 18.) 

Type: West end of Antelope Valley, Los Angeles County, 21 miles west 
of Lancaster, Calif., elevation 2,940 feet, /. M. Webber, April 11, 1946, No. 
424, Univ. Calif. Herbarium; cotype No. 423, U. S. Dept. Agr. } Plant Indus. 
Herbarium. 

Range: Extending from type locality northwest along the southern and 
western slopes of the Tehachapi Mountains to at least Monolith, Calif. 

The form is dedicated to the writer's father, Herbert John Webber, who 
noticed the distinguishing characteristics of the form. 

DEHISCENT-FRUITED YUCCAS 

10. Yucca whipplei 

(Our-Lord's-Candle, chaparral yucca, quixote yucca) 

Plant simple, solitary, dying when fruit has formed Y. whipplei. 

Plant caespitose or forming a large, fairly open clump, fruiting for many years. 
Plant caespitose, very dense and compact. 

Secondary rosettes formed in seedling stage and in axis of leaves, usually a 

large clump of many rosettes, and several flower stalks each spring 

Y. whipplei var. caespitosa. 



16 Latin description translated from English by John Thomas Howell, California Acad- 
emy of Sciences: 

Planta globum densum usque ad 150 sq. m. tegentem faciens, 7.5 fasciculis foliorum et 
1.8 caudicibus basilaribus in 1 sq. m., caulibus usque ad 3.0 — 6.5 m. (4.0. m.) altis. 



YUCCAS OF THE SOUTHWEST 33 

Secondary rosettes formed after flowering near base of old flower stalks, 
usually a comparatively small clump and only one flower stalk each spring. 

Y. WHIPPLEI var. INTERMEDIA. 

Plant a large, fairly open clump; secondary rosettes mainly developed from 

rhizomes Y. whipplei var. percursa. 

Yucca whipplei Torr., in Emory, U. S. and Mex. Bound. Bot., p. 222. 1859. 

PYucca calif ornica Groenland, Rev. Hort. (Paris), p. 434. Aug. 16, 1858. 
Y. gramini folia Wood, Acad. Nat. Sci. Phila. Proc, p. 167. 1868. 
Hesperoyucca whipplei Baker, Kew Roy. Bot. Gard. Bui. Misc. Inform. 

5: 8. 1892. 
Yucca whipplei var. parishii Jones, West. Bot. Contrib. 15: 59. 1929. 
Y. whipplei subsp. typica Haines, Madrono 6: 43. 1941. 
Y. whipplei subsp. parishii Haines, Madrono, loc. cit. 
Y. newberryi McKelvey, Yuccas Southwest. U. S. 2: 49- 1947. 
Y. whipplei McKelvey, Yuccas Southwest. U. S. 2: 42. 1947. (Solitary 

plants.) 

Plant acaulescent, simple, dying after fruit has been formed; leaf blade 25 
to 115 (65) cm. long, 5 to 35 (18) mm. wide, linear or rarely narrowly 
lanceolate, plano-convex to subtriquetrous, or keeled on both surfaces, finely 
striate, rigid and swordlike to flexible and frequently falcate, commonly glau- 
cous; leaf margin thin, corneous, more or less serrulate; scape 0.9 to 4.5 m. 
tall; panicle 0.5 to 3.5 (1.5) m. tall; 0.2 to 0.9 (0.45) m. in diameter, cylin- 
drical or somewhat slender-ellipsoidal, densely flowered; flowers usually ex- 
panding broadly, white or tinged with purple; perianth segments broadly 
lanceolate, the sepals and petals nearly equal in size, 30 to 65 mm. long, 8 to 
25 mm. wide; filaments usually exceeding the pistil, not united at base and 
without well-differentiated clavate tip, 8 to 20 mm. long; ovary stout, 6 to 10 
mm. in diameter, 8 to 12 mm. long, oblately ellipsoidal or obovoid, terminating 
abruptly with squarish shoulders, often depressed at apex; style short and 
slender; stigma capitate, with elongated, translucent papillae; capsule 30 to 
50 mm. long, 15 to 30 mm. in diameter, obovoid, acute or short-mucronate, 
rarely constricted, dehiscence loculicidal by splitting along secondary fissures of 
the capsule; seed 6 to 7 by 8 mm., flat, thin, smooth, dull black, without mar- 
ginal wing. (PI. 19.) 

Type: San Pasqual, San Diego County, Calif., Schott, Torrey Herbarium, 
N. Y. Bot. Gard. 

Range: Baja California boundary, northwest through San Diego, Orange, 
Riverside, San Bernardino, Los Angeles, Ventura, Santa Barbara, Kern, San 
Luis Obispo, and Tulare Counties to south-central Monterey and southeastern 
San Benito Counties, Calif. In northwestern Arizona, it occurs in scattered 
patches along the Colorado River from approximately Pierce Ferry east to the 
mouth of Diamond Creek, Mohave County. Although the species is usually 
scattered on mountain slopes and mesas of chamisal, chaparral, and desert 
woodlands from near sea level to 4,500 feet elevation, it occasionally extends 
well into the montane forest up to 7,500 feet. The species usually flowers 
between mid-March and the end of May. 



34 AGR. MONOGRAPH 17, U. S. DEPT. OF AGRICULTURE 

Yucca whipplei variety caespitosa 

Yucca wbipplei var. caespitosa Jones, West. Bot. Contrib. 15: 59. 1929. 

Yucca wbipplei subsp. caespitosa Haines, Madrono 6: 43. 1942. 

Y. whipplei McKelvey, Yuccas Southwest. U. S. 2: 24 (clumped plants, in 

part). 1947. . 

Variety caespitosa differs from the species in the following respects: It is 
compound and ultimately forms a large, dense clump of numerous, clustered 
rosettes; it lives for many years and usually produces several flower stalks each 
spring; and its leaf blade is slightly shorter (58 mm.), more glaucous, and 
quite rigid. When variety caespitosa seedlings are 6 months to 2 years old, buds 
are formed in the axils of the leaves. These buds soon give rise to numerous 
secondary rosettes. Development of the secondary rosettes is accompanied 
by the formation of short, lobelike stems or branches. The branches are al- 
ways on the surface of the ground, and the stem cannot be observed unless 
the rosettes are split open or the leaves removed. 

Type: Cactus Flat in Cushenbury Canyon, San Bernardino County, Calif., 
M. E. Jones, May 12, 1926; Pomona College No. 122326, California Acad. 
Sci. No. 150109. 

Range: Southern and western slopes of the Mojave Desert, from the 
vicinity of Baldwin Lake, San Bernardino County, to the region of Walker 
Pass, northwestern Kern County, Calif. It is found in desert juniper wood- 
lands and shrubs at 1,800 to 4,000 feet elevation. 

Yucca whipplei variety intermedia 

Yucca wbipplei var. intermedia (Haines) comb. nov. 

Yucca wbipplei subsp. intermedia Haines, Madrono 6: 43. 1942. 

Y. wbipplei var. caespitosa Jones, West. Bot. Contrib. 15: 59 (in part). 

1929. 
Y. wbipplei McKelvey, Yuccas Southwest. U. S. 2: 24 (clumped plants, 

in part). 1947. 

Variety intermedia is similar to caespitosa in that it lives for many years and 
forms a dense clump of rosettes. The clumps of intermedia, however, are 
comparatively small, and the plant rarely forms more than one inflorescence 
in a single season. Although the leaf blade is similar in size to that of the 
species, it is possibly somewhat more rigid and glaucous than that of variety 
caespitosa. In variety intermedia secondary rosettes are formed near the base 
of old flower stalks from apparently adventitious axillary buds, which develop 
below the dying portion of rosette and flowering stalk. Following secondary 
rosette formation, short, lateral branches are formed similar to those in variety 
caespitosa. In variety intermedia, however, the old mother rosette rarely pro- 
duces more than one or two secondary rosettes and branches. 

Type: Malibu Lake, Santa Monica Mountains, A. L. Haines, Univ. Calif., 
Los Angeles Herbarium. 

Range: Santa Monica and Santa Susana Mountains of Los Angeles and 
Ventura Counties, Calif., from near sea level to 2,000 feet elevation. 



YUCCAS OF THE SOUTHWEST 35 

Yucca whipplet variety percursa 

Yucca wbipplei var. percursa (Haines) comb. nov. 

Yucca whipplet subsp. percursa Haines, Madrono 6: 43. 1942. 
Y. wbipplei McKelvey, Yuccas Southwest. U. S. 2: 24 (clumped plants, 
in part). 1947. 

Variety percursa is similar to the preceding two varieties in that upon reach- 
ing maturity it forms clumps, flowers, and fruits for many years. Variety 
percursa differs from the other varieties, however, in that its clumps are larger 
and rather open and composed of distinct, scattered rosettes and rosette clusters. 
It differs from all other members of the species in that its leaf blades are 
shorter (25 to 90 (48) mm.) and somewhat more rigid. In variety percursa, 
secondary rosettes are mainly of rhizomatous origin, which results in a rather 
open, large clump. Following flowering, however, the secondary rosettes 
occasionally produce adventitious axillary rosettes, which add to the density and 
confused nature of the clump. 

Type: Cochuma Mountain in San Rafael Mountains of Santa Barbara 
County, Calif., A. L. Haines, Univ. Calif., Los Angeles Herbarium. 

Range: San Rafael, Santa Ynez, and Santa Lucia Mountains of Santa Bar- 
bara and Monterey Counties, Calif., on mountain slopes and mesas of sage and 
chaparral, from near sea level to about 2,000 feet elevation. 

Discussion of Varietal Forms 

In a study of variations and growth habits of Yucca wbipplei, Haines (15) 
segregated five subspecies. Three of these subspecies were the varieties de- 
scribed above. Haines fourth and fifth subspecies, typica and parishii (var. 
parishii Jones), respectively, were segregated from the species as here denned 
by the following characters: For subspecies typica flower stalk 5 to 13 feet high, 
1 to 4.5 inches in diameter; panicle 1.5 to 7 feet long, 0.5 to 2 feet in di- 
ameter; and for subspecies parishii flower stalk 7 to 20 feet high, 1.5 to 6 
inches in diameter; panicle 3 to 11.5 feet long, 1 to 3 feet in diameter. Accord- 
ing to Haines, the five types were possibly geographic races, segregated by 
geographic barriers; he thus treated them as subspecies. 

McKelvey (25) called attention to the fact that, possibly owing to environ- 
mental differences, all yuccas are quite variable in size, and she cited several 
localities in which the size relationships of subspecies parishii and subspecies 
typica do not exist. She therefore concluded that the segregation of these two 
types was unjustified. The field findings of the writer confirm those of 
McKelvey. 

McKelvey believed that three of Haines' attributes of subspecies intermedia 
suggested an intermingling of a caespitose and a monocarpic form. These 
characteristics (25, p. 39) were as follows: (1) ". . . branching 'by means ol 
axillary buds to form short lateral branchlets' {intermedia) appears inter- 
mediate between branching 'on the surface of the ground' (caespitosa) and 
non-branching (typica and parishii)"; (2) intermedia, a caespitose form, is 
occasionally solitary; and (3) intermedia, with a group of rosettes, forms only 
one inflorescence a season. McKelvey concluded, "It seems probable . . . that, 
over the entire area between these northern [caespitosa'] and southern [typica 
and parishii'] forms, and extending into the ranges of each, a race of unstable 
and intermediate plants should exist." 



36 AGR. MONOGRAPH 17, U. S. DEPT. OF AGRICULTURE 

It is the writer's opinion that there is little or no relationship between the 

methods of branching in varieties caespitosa and intermedia. The branching of 
variety intermedia appears to be purely adventitious, possibly stimulated by 
flowering. The method is characteristic of the majority of yuccas, and it is 
not uncommon in long-stemmed forms to find the branches (or new leaf heads) 
developing below the fresh leaves of the old dying leaf head and flower stalk. 
Furthermore, occasionally such forms fail to flower, in which case no secondary 
leaf heads or branches are formed. In variety caespitosa, the branches are 
formed from buds developed at a definite time and place. In the majority of 
plants of variety caespitosa and variety intermedia there is no appreciable dif- 
ference in height of branching. Such height differences as do occur are 
undoubtedly caused by the relative age and crowded condition of the plant. 
The secondary rosettes of mother-seedling rosettes of both varieties are approxi- 
mately at soil level. 

Haines' statement that variety intermedia was occasionally solitary probably 
indicated that there was hybridization, but not necessarily that all plants of Y. 
whip pi ei var. intermedia were hybrids nor that variety caespitosa was involved 
in its production. If variety ititermedia were a hybrid between variety caespitosa 
and a solitary form, one would expect caespitosa to occur as well as the solitary 
form. That variety intermedia is not a hybrid is indicated by its occurrence in 
pure concentrations. 

Although the writer is unable to suggest why Y. whipplei var. intermedia 
forms only one inflorescence a year, he does not believe it indicates hybridiza- 
tion. 

Since McKelvey's knowledge of varieties percursa and caespitosa was mainly 
limited to herbarium material, she considered herself unqualified to express 
an opinion upon their distinctions. She pointed out, however, that field and 
garden experience suggested that soil texture (rocky or open) might have 
some bearing upon the length of offset and, hence, density of plant. Although 
such response to soil texture is unquestionable, nursery and field evidence in- 
dicates that the methods of branching and the main characteristics of the 
branches are inherited. 

Haines (15, p. 36) stated: "The only points at which any of these races are 
known to merge and exhibit intermediates lies in the Cajon Pass area where 
hybridization is apparently frequent between the solitary montane form and 
the caespitose desert form." In this region and along the southwestern slopes 
of the San Gabriel Mountains, the plants appear to form a variable group of 
intermingled caespitose and monocarpic forms. Although typical Y. whip pie/ 
and variety caespitosa occur, intermediate plants are more common. The inter- 
mediate plants are much more vigorous than either Y. whipplei or its variety 
caespitosa and exhibit fewer rosettes than caespitosa. Their leaves are larger, 
stiffer, and more glaucous than either the species or variety. The most dis- 
tinctive character of these apparent hybrids are the occasional occurrence of 
short rhizomes in older dense plants similar to those of variety percursa, and of 
short (8 to 20 cm.) thin-stemmed axillary rosettes in comparatively young 
seedlings. The behavior of nursery seedlings from the Cajon Pass area verifies 
the hybrid nature of these plants. 

Relative to the capsules of Y. whipplei and Y. peninsularis McKelvey (a 
Baja California species), McKelvey (25, p. 22) stated that they show " a well- 
developed false septum which consists of a conspicuous fibrous fringe borne 
on the midrib of the carpel and which projects well into the cell-cavity and has 
its distal margin lying between two conspicuous tough plates of tissue that 



YUCCAS OF THE SOUTHWEST 37 

appear to be laminar outgrowths from the two placental lines on the inner 
angle of the carpel; in the open capsule these two plates give a 'winged' ap- 
pearance to the placenta." In a solitary form of Y. wbipplei in northwestern 
Arizona, McKelvey found that the placental wing was weak or possibly rudi- 
mentary. McKelvey also found that these plants resembled the caespitose, 
perennial form of Y. wbipplei in foliage, in fragility of inflorescence, and in 
small fragile, lightweight, eventually widespreading capsule with pale tan 
epidermis. The rudimentary nature of the placental wing and the monocarpic 
behavior in conjunction with the characteristic caespitose form led McKelvey 
to segregate these plants from Y. whip pi ei as Y. newberryi. 

Yucca newberryi (23) is reported on the southern side of the Colorado 
River from approximately the west end of the Grand Canyon to Lake Mead, 
Ariz. On several occasions the writer has seen a few scattered plants of the 
form approximately 3 miles northeast of Pierces Ferry and in the vicinity of 
New Water Point. Since, however, he has not seen the plants in flower nor 
carefully observed their capsular structure, he is unqualified to express an opin- 
ion upon these organs. It is the writer's impression, however, that Y. new- 
berryi is separated on too variable and minor characters. In the vicinity of 
Pierces Ferry, the leaves of Y. wbipplei average about 65 cm. long, 3 cm. wide, 
and are subtriquetrous or somewhat 4-sided (keeled distinctly below and slightly 
above); the inflorescence (old) is approximately 3 m. tall; and the capsules 
rather small. All of these characters are well within the range of variations 
exhibited by Y. wbipplei. In the majority of yuccas, capsule texture, color, and 
degree of expansion, as well as size and shape, are very variable. The capsules 
of Y. elata are fragile and papery in the vicinity of Salome, Ariz.; are white 
and comparatively smooth in White Sands National Monument, N. Mex., and 
the dorsal suture at the apex of the capsule commonly separates, forming a 
pocket, in the vicinity of Pecos, Tex. In the vicinity of Tucumcari, N. Mex., 
the capsules of Y. glauca allies vary from those that never open or expand to 
ones that expand fully or nearly flat. 

Although members of the Y. wbipplei alliance are associated with, or found 
in close proximity to Y. brevifolia, Y. scbidigera, and Y. baccata, no evidence 
of hybrids between it and these species has been found. 

Among our southwestern yuccas, members of the Y. wbipplei group are 
unusual in several ways. It is the only species that produces a definite bulb 
(seedling stage) or that has a capitate stigma and strictly loculicidally dehis- 
cent capsules. The inflorescence of Y. wbipplei in several parts of its range, 
mainly at higher elevations, is by far the largest and most attractive of yuccas, 
and its flowers are much more fragrant than those of Y. carnerosana or Y. 
faxoniana, the only other fragrant yuccas in the Southwest. Y. wbipplei is the 
only southwestern yucca that produces a single rosette, a single inflorescence, 
and then dies. Evidence indicates that the species completes its entire life in 
4 to 6 years, whereas all other yuccas live indefinitely and usually flower each 
year after reaching maturity. Furthermore, the failure of Y. wbipplei to pro- 
duce secondary rosettes means that it is the only yucca that depends entirely 
upon seed for propagation and distribution. The variety caespitose is the only 
yucca that produces axillary rosettes at a definite period of life and in a definite 
axil. As the axillary rosettes are a method of propagation, the Y. wbipplei 
group is the only species group having three distinct types of propagation; that 
is, seed, axillary bulbose- rosettes, and rhizomatous- rosettes. 



38 AGR. MONOGRAPH 17, U. S. DEPT. OF AGRICULTURE 

11. Yucca rostrata 

(Big Bend yucca) 

Yucca rostrata Engelm. ex Trel., Mo. Bot. Gard. Ann. Rpt. 13: 68. 1902. 
Yucca rostrata f. Integra Trel., Mo. Bot. Gard. Ann. Rpt. 22: 102. 1911. 

Plant arborescent, rather symmetrical, with 1 to 5 (1-9) shoots; stem 1 to 
4.5 (2.3) m. tall, 18 to 32 (27.2) cm. in diameter, trunklike, erect, with to 
3 (1.5) short assurgent branches toward the top; stem and branches covered 
with reflexed dead leaves; leaf head symmetrical, large and elongated; leaves 
numerous, the blade 42 to 61 (49.5) cm. long, 12 to 17 (14) mm. wide, 
linear, or somewhat broader toward center, flat, concavo-convex or keeled on 
both surfaces, striate, thin, flexible, glaucous, pungent; leaf margin corneous, yel- 
low, hyaline, minutely denticulate; scape 30 to 70 (54) cm. long, stout, glabrous 
or evanescently pubescent; panicle to 20 (15.6) cm. above foliage, 35 to 70 
(56.6) cm. long, narrowly ellipsoidal to somewhat ovoid, densely flowered, 
with 28 to 40 branchlets, 2 to 35 cm. long; flowers white, globose to campanu- 
late, fully expanding at anthesis, with umbonate base; perianth segments nar- 
rowly oblong, sharply acuminate, conspicuously veined; sepals 42 to 47 (44) 
mm. long, 12 to 16 (14) mm. wide; petals 42 to 50 (45.2) mm. long, 11 to 
20 (15.3) mm. wide; filaments 17 to 20 (19) mm. long, with fairly erect 
clavate tip, pistil 25 to 35 (27.5) mm. long; ovary 4 to 6 (5) mm. in di- 
ameter, usually tapering into style, or rarely somewhat abruptly terminating in 
style; style 6 to 14 (8) mm. long, attenuate, white; capsule 35 to 70 mm. long, 
18 to 25 mm. in diameter, ellipsoidal or somewhat ovoid, with long-attenuate 
or rarely short-attenuate beak and obconical pedicel, rarely constricted; beak of 
dry capsule flaring, with long valve points outcurved and twisted; seed 4 to 
5 by 6 to 7 mm., flat, thin, dull black, without marginal wing. (PL 20.) 

Type: Monclova, Mexico, E. Palmer 1880; Mo. Bot. Gard. No. 148694. 

Range: Northern Mexico, from northern Chihuahua to the Sabinas Valley 
in eastern Cochuila. In the United States it is limited to south-central and 
southeastern Brewster County, Tex. Although the species is usually rather 
thinly scattered on mountain slopes and bajadas at 2,400 feet elevation, it 
occasionally occurs in quite dense concentrations of several acres in area. The 
species usually flowers between March and April. 

Yucca rostrata is one of our most attractive and graceful yuccas. The heads 
of leaves are large, very symmetrical, and vary from nearly globose to cylin- 
drical in shape. Its numerous, rather glaucous leaves, frequently twisted so 
that the upper and lower side are reversed toward the middle, radiate from 
the head center. The old, dead leaves closely reflexed against the stem in def- 
inite growth cycles form a persistent straw-colored, thatchlike covering over the 
majority of the stem. 

Although the leaf margins of most plants, bear minute teeth, the character 
appears somewhat unstable. Rarely, the denticulations are lacking. The entirely 
smooth-edged plants have been referred to as Y. rostrata forma Integra Trel. 

Yucca rostrata is commonly admixed with Y. carnerosana and Y. torreyi, 
and possibly in some sections of its range may be associated with Y. elata. In 
habitat Y. rostrata and Y. elata are quite similar, and at a distance they are 
readily mistaken for one another. No evidence of hybridization between Y. 
rostrata and other yuccas has been observed. 



YUCCAS OF THE SOUTHWEST 39 

12. Yucca tbompsoniana 

(Trans-Pecos yucca) 

Yucca tbompsoniana Trel., Mo. Bot. Gard. Ann. ,Rpt. 22: 101. 1911. 

Yucca rostrata Engelm. ex Trel., Mo. Bot. Gard. Ann. Rpt. 13: 68 (in 
part). 1902. 

Plant arborescent, frequently asymmetrical or rarely symmetrical, with 1 to 
3 (1.3) shoots, stems 0.7 to 2.6 (1.4) m. tall, 12 to 15 (14) cm. in diameter, 
trunklike, erect with comparatively long, assurgent or diffused branches; leaf 
heads frequently asymmetrical, rather small and composed of comparatively 
few leaves; leaf blade 18 to 30 (21.7) cm. long, 7 to 12 (8.3) mm. wide, 
linear or somewhat broader toward center, flat or concavo-convex to piano- 
keeled, striate, thin and flexible; leaf margin corneous, yellow or brownish, 
minutely denticulate; scape 38 to 68 (55.6) cm. long, stout, glabrous or evane- 
scently pubescent; panicle 11 to 19 (15.6) cm. above foliage, 52 to 82 cm. 
long, narrowly ellipsoidal to somewhat ovoid, densely flowered with 20 to 34 
branchlets, 2 to 22 cm. long (average of 10 longest, 17 cm.); flowers white, 
globose to campanulate, spreading flat at an thesis, with umbonate base; perianth 
segments narrowly oblong, sharply acuminate, conspicuously veined; sepals 35 
to 67 (46.9) mm. long, 12 to 26 (15.9) mm. wide; petals 40 to 66 (49.9) mm. 
long, 11 to 35 (19.8) mm. wide; filaments 17 to 33 (23.0) mm. long, with 
fairly erect clavate tip; pistil 25 to 38 (31.1) mm. long; ovary 4 to 6 (4.4) 
mm. in diameter, slender, usually tapering into style, or rarely somewhat ab- 
ruptly terminating in style; style 6 to 18 (14) mm. long, attenuate, white; cap- 
sule 35 to 70 mm. long, 20 to 25 mm. in diameter, ellipsoidal or somewhat 
ovoid, with long attenuate or rarely short-attenuate beak and obconical pedicel, 
rarely constricted, beak of dry capsule flaring, with long reflexed, often twisted, 
valve points; seed 5 to 6 by 6 to 7 mm., flat, thin, dull black, without marginal 
wing. (PI. 21.) 

Type: Bufatello, near Presidio del Norte, along the Rio Grande, Mexico; 
J. M. Biglow, August 1852, N. Y. Bot. Gard. 

Range: North and central Brewster County, east-central and southern Pecos 
County, throughout Terrell County, and western half of Crockett County, Tex. 
The species is usually thinly scattered on exposed rocky knolls and slopes at 
900 to 4,400 feet elevation. Fairly good, but small concentrations occur \6 
miles east of Marathon, Tex., in the State game preserve; 28 miles south of 
Marathon, in the Texas Rio Grande Big Bend area; and 10 to 12 miles east 
of Fort Stockton, Tex. The species usually flowers between April 15 and May 
30. 

In many respects Yucca tbompsoniana appears to be a dwarf form of Y. 
rostrata, which occurs slightly south and southwest of its range. It cannot be 
considered a nanate form of Y. rostrata, however, since several of its organs arc 
equal in size or larger than those of Y. rostrata. Its few branches are longer 
and more diffused. Frequently in old plants the main stem (trunk) appears 
too slender and weak to support the relatively long spreading branches, and 
the branches themselves often look as if they would fall of their own weight. 
The leaves and leaf heads of Y. tbompsoniana are much smaller than those of 
Y. rostrata, and the older heads of Y. tbompsoniana are frequently ragged in 
appearance. The inflorescences and flowers of both species are about equal 
in size. 



40 AGR. MONOGRAPH 17, U. S. DEPT. OF AGRICULTURE 

On several occasions it has been brought to the writer's attention that the 
persistent old leaves, which cover the majority of stems of Y. ro strata, may be 
a delimiting character, aiding in separating Y. rostrata and Y. thompsoniana. It 
is his opinion that this feature is characteristic of both species. In the majority 
of cases Y. rostrata is well protected from wind erosion by mountains and fairly 
heavy growths of other plants, but Y. thompsoniana is usually exposed on more 
or less gentle slopes, which have very scanty vegetation as a result of sheep 
grazing. Consequently, the majority of Y. rostrata stems are covered with 
a thatchlike covering of dried-out leaves, whereas the dead leaves of Y. thomp- 
wniana are usually worn away by erosion. Protected plants of both forms 
exhibit persistent leaves, which completely cover the stems. 

Scattered on a hillside, 2 miles south of Leon Powell Oil Field, northwest 
of Ozona, Crockett County, Tex., are small weak yuccas that resemble Y. 
thompsoniana. The plants have 1 to 4 stems, which, upon reaching 0.6 to 1.0 
m. tall, bend to the ground, as if they can no longer support their weight. 
Decumbent stems up to 1.5 m. long are not uncommon. The plants exhibit 
very little clumping, and branching is mainly well above the ground. The 
writer has not seen either inflorescence or flowers of these plants. As the 
general habit of these plants differs considerably from that of the apparent 
hybrids discussed below, they are very likely dwarf or weak plants of Y. 
thompsoniana in its extreme north and east range. 

Apparent Hybrids of Yucca thompsoniana and Yucca reverchoni 

In the vicinity of Bakersfield and south toward Sanderson, Tex., apparent 
hybrids between Yucca thompsoniana and Y. reverchoni are common. These 
plants resemble Y. thompsoniana in their rather tall (0.5 to 1.0 m.), branched 
stems, and long (1.0 to 1.3 m.) flowering stalk. They resemble Y , reverchoni 
in their decidedly clumped habit and narrow panicle. The clumps are from 
1.0 to 1.8 m. tall, of approximately the same diameter, and are composed of 
7 to 20 heads of leaves, which are equally derived from subterranean and 
aerial stems. The writer has not seen the flowers of these plants. 

Yucca thompsoniana is frequently associated with Y. torreyi, occasionally 
with Y. baccata and Y. reverchoni, and rarely with Y. elaia. The writer has not 
found it admixed with Y. rostrata, which grows somewhat farther south. No 
further evidence of hybrids between Y. thompsoniana and other yuccas has been 
observed. 

13. Yucca rupicola 

Yucca rupicola Scheele, Linnaea 23: 143. 1850. 

Yucca rupicola tortifolia Engelm., Acad. Sci. St. Louis. Trans. 3: 49- 

1878. 
Y. rupicola Trel., Mo. Bot. Gard. Ann. ppt. 13: 67 (in part). 1902. 

Plant acaulescent, solitary but soon becoming an open clump of scattered leaf 
heads; clump 1 to 2 sq. m. in area, with 6 to 15 heads of leaves; leaf heads 
large, spreading, with few leaves; leaf blade 30 to 58 (43.3) cm. long, 20 to 
40 (30) mm. wide, very broad toward middle, concaved or flat, but oblique 
and undulate or twisted, slightly striate, flaccid, dark green, pungent; leaf mar- 
gin hyaline reddish brown or occasionally yellow, minutely denticulate; scape 



YUCCAS OF THE SOUTHWEST 41 

36 to 152 (89.9) cm. long, slender, 15 to 20 mm. in diameter; glabrous to 
lightly floccose; panicle 24 to 48 (44.4) cm. above foliage, 31 to 100 (59.5) 
cm. long, narrowly ovoid or narrowly pyramidal, few-flowered, with 8 to 16 
branchlets; branchlet 1 to 13 cm. long (the longest, 6 to 13 cm.) ; flowers 
mainly campanulate, pendent, and expanding but little, rarely somewhat globose 
and open; white or greenish white; perianth segments ovate, sharply acuminate; 
sepals 38 to 68 (50.1) mm. long; 15 to 24 (18.8) mm. wide; petals 38 to 
69 (51.2) mm. long. 19 to 30 (24.5) mm. wide; filaments 18 to 32 (25.9) 
mm. long; pistil 29 to 45 (40.3) mm. long; ovary 4 to 6 (4.9) mm. in di- 
ameter, tapering into style or somewhat abruptly terminating in style; style 12 
to 20 (16.1) mm. long, attenuate, white or greenish; capsule 38 to 54 mm. 
long, 20 to 30 mm. in diameter, ellipsoidal or somewhat cylindrical with 
long-attenuate, or short-attenuate beak and obconical pedicel, rarely constricted; 
beak of dry capsule flaring with valve points generally outcurved and twisted; 
seed 6 to 8 by 7 to 8 mm., flat, thin, dull black, without or with very narrow 
marginal wing. (PI. 22.) 

Type: South-central Texas, Undheimer, 1845, Trecul, 1848-49. 

Range: Extending from Palo Pinto and Tarrant Counties, Tex., southwest 
to the Pecos and Rio Grande. It occurs on limestone ledges and on grass-covered 
plains of dense brush and open woodlands at 1,500 to 2,900 feet elevation. 
The plants are thinly scattered and only rarely occur in small patches. The 
writer's studies of the species were made in the vicinities of Junction, Kimble 
County, Kerrville, Kerr County, and Fredericksburg, Gillespie County, Tex., 
where it occurs abundantly. The majority of its reported southwestern range 
is occupied by apparent hybrids between the species and Yucca reverchoni. 
It flowers between mid-May and mid- June. 

Trelease (31) suggested the name Y. rupicola edentata for a smooth-edged, 
or entire, leaf form of Y. rupicola collected at Cedar Hill, Tex., by Julien 
Reverchon, June 1903. The writer has not found any entire-leaved forms of the 
species. 

McKelvey (25, p. 15) described Y. rupicola as "Plants with single head of 
leaves or in small clump with 2-6 heads; matured leaf-blade twisted, very 
concave . . . , with wavy, dark orange- red or red-brown margins, ..." and 
gives the range of the species as "central Texas; especially prevalent over the 
eastern end of the Edwards Plateau ..." North and somewhat east of the 
latter range, McKelvey segregated Y. pallida, mainly by its: "Plants usually 
with 10-30 heads of leaves in one clump; mature leaf-blade straight, flat 
except for 1.3-2.5 cm. below tip ... , with flat, bright yellow margins ..." 
Although the range of Y. pallida is mainly east of this yucca survey, the 
writer found Y. rupicola on the Edwards Plateau to have 6 to 15 heads of 
leaves, and the leaf blade flat or concaved, with occasionally yellow margins. 

14. Yucca reverchoni 

(San Angelo yucca) 

Yucca reverchoni Trel., Mo. Bot. Gard. Rpt. 22: 102. 1911. 

Yucca rupicola Scheele, Linnaea 23: 143 (in part). 1850; Trel., Mo. 
Bot. Gard. Ann. Rpt. 13: 67 (in part). 1902. 

Plant acaulescent, solitary but ultimately a dense small clump; clump 0.3 to 
1.0 m. in diameter with 1 to 25 (6.7) heads of leaves; leaf head small, with 



42 AGR. MONOGRAPH 17, U. S. DEPT, OF AGRICULTURE 

few leaves; leaf blade 25 to 55 (38) cm. long, 10 to 20 (14) mm. wide, 
linear to somewhat broader toward the center, concavo-convex, quite rigid, 
straight, light glaucous green; leaf margin hyaline yellow or occasional)- red 
or brown, minutely denticulate; scape 46 to 110 (83.84) cm. long, slender, 
glabrous to heavily floccose; panicle 25 to 42 (33.6) cm. above foliage. 36 to 
100 (58.7) cm. long, narrowly ovoid or narrowly pyramidal, with few branch- 
lets and flowers; flowers pendent, campanulate to somewhat globose, expanding 
but little at anthesis, white or greenish white; perianth segments ovate, sharply 
acuminate; sepals 38 to 60 (50.3) mm. long, 15 to 21 (18) mm. wide; petals 
38 to 59 (50.4) mm. long, 19 to 29 (24.6) mm. wide; filaments 18 to 32 
(25.7) mm. long; pistil 29 to 45 (38.0) mm. long; ovary 4 to 6 (4.8) mm. in 
diameter, tapering into style or rarely abruptly terminating in style; style 9 
to 20 (15) mm. long, attenuate, white or greenish; capsule 38 to 59 mm. long, 
18 to 31 mm. in diameter, rarely constricted, ellipsoidal with long-attenuate 
or short-attenuate beak and obconical pedicel; beak of dry capsule flaring, with 
long reflexed valve points; seed 5 to 6 by 6 to 7 mm., flat, thin, dull black, 
without marginal wing. (PI. 23.) 

Type: Twin Mountain, San Angelo, Tex., Julie?? Revercho??, May 20, 1908; 
Mo. Bot. Gard. No. 148679. 

Range: McCamey, Upton County, Tex., and extending northeast, east, and 
southeast through Reagan, Irion, Crockett, Val Verde, Sutton, and Kinney 
Counties to the vicinities of San Angelo, Tom Green County; Fort McKavett, 
Menard County; Roosevelt, Kimble County; Vance, Real County; and Uvalde, 
Uvalde County, Tex. It is usually found on rocky limestone ledges and gravelly 
plains of dense brush, at 1,200 to 2,950 feet elevation. The plants are thinly 
scattered. Probably the best plants of the species occur approximately 1.5 
miles north of Eldorado, Tex. The species flowers between the first of May 
and mid-June. Although the inflorescence is very attractive, the dense clump 
is rather untidy in appearance. 

McKelvey {25, p. 76) had difficulty in separating Yucca revercho??? from 
stemless or short-stemmed plants of Y. tho??ipso??ia?ta and concluded it was an 
intermediate race between Y. rupicola and Y. tho??ipso??ia??a. Relative to the 
race, McKelvey stated, ". . . fruit does not synchronize with foliage; . . . 
inflorescences show frequent abnormalities, several flowers united into one. 
rhachis and branchlets fasciate . . . The foliage of some, indeed of most, is 
long and rather slender, suggesting Y. rupkola influence; however, in certain 
specimens ... it is short as in Y. Tho??ipso??'ia??a but too broad for that species 
... In the field, where this race is found, plants were neither universally ac- 
aulescent . . . nor truly arborescent . . but varied from stemless to shrub- 
like, the stems prostrate to more or less erect and 0.3-0.7 m. in length; notice- 
able also was a vigorous growth which produced a denser, more untidy crown 
than is characteristic of either species." 

Apparent Hybrids of Yucca reverchoui and Yucca rupicola 

Along the Nueces and Rio Frio, in southern Uvalde, and in southern Ed- 
wards and Real Counties, Tex., apparent hybrids and hybrid segregates be- 
tween Yucca revercho??} and Y. rupkola are quite common. These plants 
resemble Y . revercho??? in their small, dense clump and their light glaucous 
green leaves; and Y. rupkola in their long (42 to 56 cm.) more or less flaccid 
leaves. They are approximately midway between the two species in leaf width 



YUCCAS OF THE SOUTHWEST 43 

(1.9 to 2.4 cm.), obliqueness, and undulation. The leal margins are hyaline 
yellow, light brown, or deep reddish brown. Between Junction and Roosevelt, 
the growth habit of the hybrid is more characteristic of Y. rupicola, whereas 
the leaves closely resemble Y. reverchoni. At Roosevelt and near McCamey, oc- 
casional hybrids occur between plants of species in the Y. rupicola and Y. 
glauca alliances. These plants mainly resemble Y. rupicola, except for small, 
greenish, globose flowers with a short style, or resemble Y. glauca, except for 
nonfiliferous leaves that are frequently somewhat denticulate and occasionally 
denticulately keeled on the under surface. 

Possible hybrids between Y. reverchoni and Y. thotnpsoniana are discussed 
under Y. thompsoniana (p. 40). 

15. Yucca neomexicana 

(New Mexico yucca) 

Yucca neomexicana Woot. & Standi., U. S. Natl. Herbarium Contrib. 16: 115. 

1913. 
Yucca harrimaniae McKelvey, Yuccas Southwest. U. S. 2: 139 (in part). 

1947. 
Y. coloma Andrews, Cat. Rockmont Nursery 1926: 22. 

Plant acaulescent, forming an open clump of 5 to 15 heads of leaves, 35 to 
60 cm. apart, or rarely somewhat caespitose; head of leaves small, symmetrical; 
leaf blade 15 to 46 (20.5) cm. long, 7 to 20 (13) mm. wide, linear to lan- 
ceolate, concavo-convex, thin, rigidly spreading, but easily flexible, deep green, 
rather glaucous, pungent; leaf margin white or at length moderately nliferous, 
the fiber fine, straight; scape 13 to 44 (23) cm. long, glabrous; inflorescence 
mainly a simple raceme or rarely with a few short branchlets; raceme to 20 
(5) cm. above foliage, 45 to 70 (51) cm. long, rather openly flowered; 
flowers broadly campanulate, expanding but little, pendent, white or somewhat 
greenish white, commonly tinged wth pink or purple; perianth segments broad, 
obtuse; sepals 44 to 63 (53) mm. long, 19 to 31 (25.5) mm. wide; petals 42 
to 62 (51.5) mm. long, 24 to 33 (29.5) mm. wide; filaments 17 to 28 (22) 
mm. long, very hirsute; pistil 27 to 36 (30) mm. long; ovary 5 to 7 (6) 
mm. in diameter, white or rarely pale green, abruptly terminating in style 
or shouldered at apex, with staminal depressions wanting or very slightly, and 
deep carpel sutures; style 9 to 13 (10.5) mm. long, slightly to heavily swollen 
toward middle, pale green or rarely white; capsule 30 to 40 mm. long, 20 to 
25 mm. in diameter, cylindrical with short attenuate beak, usually deeply con- 
stricted toward center and flaring open on drying; seed small, thin, very narrow 
marginal wing. (Pis. 24, 25, 29 ,D, and 29,£.) 

Type: Volcanic hill about half a mile north of Des Moines, Union County, 
N. Mex., P. C. Standley 6208, June 11, 1911, U. S. Natl. Herbarium Nos. 
285240 and 685238. 

Range: Extreme northeastern corner of Union County, N. Mex., adjacent 
northwestern corner of Cimarron County, Okla., and adjoining corners of Las 
Animas and Baca Counties, Colo. Similar plants, not completely identified but 
very likely the same species, occur in Mesa, Montrose, San Miguel, Ouray, and 
Gunnison Counties, Colo. Yucca neomexicana occurs mainly in open wood- 
land at 3,900 to 8,000 feet elevation. It is thinly scattered and rarely spreads 
into the lower grasslands or chaparral. In woodland areas it is usually found 



44 



AGR. MONOGRAPH 17, U. S. DEPT. OF AGRICULTURE 



on exposed rocky ledges, where its rhizomes wind in and out of the narrow, soil- 
filled fissures of the ledges. Aimlessly dispersed along the rhizomes are soli- 
tary rosettes, forming a very open irregularly shaped clump. When the plants 
are found in the adjacent grasslands, the rhizomes are somewhat shorter and 
the rosettes more uniformly distributed. The clumps of such plants are quite 
dense and rather uniform in shape. The species usually flowers between the 
middle of May and the last of June. 

In its New Mexico-Oklahoma range Y. neomexlcana is quite uniform, and 
its leaves, inflorescence, height of flowering, and other plant characters deviate 
very little from the averages given in the species description. On the other 
hand, apparently because of extreme differences in elevation, the plants in 
western Colorado are quite variable. They are somewhat more clumped than 
the plants in New Mexico and Oklahoma, and the leaves are longer, more 
spatulate, and vary from deeply to rather shallowly concave. Near Sapinero, 
Gunnison County, Colo., at approximately 7,600 feet elevation, the plants 
have comparatively few leaves, which are 6 to 10 (7.3) mm. wide, 14 to 34 
(23.7) cm. long, and deeply concaved. In the vicinity of Cimarron, Montrose 
County, at 5,900 feet elevation, the rosettes are larger, with numerous leaves, 
which are 12 to 15 (13.7) mm. wide, 43 to 46 (44.7) cm. long, and are 
rather shallowly concave. 

Plants identified as Y. harrimaniae Trel., which the writer believes to be 
Y. neomexicana, were reported by McKelvey (23) in the vicinity of La Vita, 
Huerfano County, Colo., and by Standley 17 in Apache County, Ariz. 

The writer confirms McKelvey's (23) findings that Y. neomexicana does not 
occur at the type locality. Although Y. glauca is fairly abundant in the vicinity, 
the nearest plants of Y. neomexicana were found to be some 35 miles east in 
the New Mexico-Oklahoma range of the species. 

Although Y. neomexicana is frequently admixed with Y . glauca and Y. 
baccata, and occasionally with Y. baileyi, no evidence of hybridization with 
these species has been observed. 

16. Yucca gilbertiana 

(Salt Lake Desert yucca) 

Yucca gilbertiana (Trel.) Rydb., Flora Rocky Mountains and Adjacent Plains, 
pp. 170, 1061. 1917. 

Yucca harrimaniae var. gilbertiana Trel., Mo. Bot. Gard. Ann. Rpt. 18: 

225. 1907. 
Y. harrimaniae Trel., Mo. Bot. Gard. Ann. Rpt. 13: 59 (in part). 1902; 

McKelvey, Yuccas Southwest. U. S. 2: 139 (in part). 1947. 

Plant acaulescent, forming a dense, small clump of 3 to 22 (6.5) heads of 
leaves, covering an area of 0.25 to 1.5 (0.8) m. in diameter; head of leaves 
asymmetrical or rarely small and symmetrical; leaf blade 12 to 50 (28.4) cm. 
long, 18 to 43 (30) mm. wide, lanceolate to spatulate-lanceolate, concavo- 
convex, deeply striate, rather thick and rigid, pale green, pungent; leaf margin 
white or brown, at length filiferous; fiber coarse and curly; scape 10 to 14 
(11.4) cm. long, inflorescence mainly a simple raceme or rarely wi:h few short 
branchlers; raceme extending 10 to 20 (16.8) cm. within foliage, 35 to 70 



17 P. C. Standley 7314, July 28, 1911, Deep Canyon, Navajo Indian Reservation, north 
end Carrizo Mountains, Apache County, Ariz., U. S. Natl. Herbarium No. 686280. 



YUCCAS OF THE SOUTHWEST 45 

(44.4) cm. long, densely flowered; bract large and persistent; flowers broadly 
campanulate, expanding but little at an thesis, pendent, yellow or greenish 
yellow, commonly tinged with purple; perianth segments broad; sepals 41 to 60 
(46.9) mm. long, 16 to 26 (19.9) mm. wide; petals 40 to 58 (47.3) mm. 
long, 21 to 34 (27.4) mm. wide; filaments 16 to 22 (18.6) mm. long; pistil 
30 to 38 (32.9) mm. long; ovary 4 to 8 (5.9) mm. in diameter, pale green 
with pronounced carpel sutures and faint staminal depressions; style 9 to 11 
(9.8) mm. long, swollen to diameter of 3 to 6 (4.4) mm. toward center, 
very pale green or white; capsule 35 to 45 mm. long, 22 to 30 mm. in 
diameter, cylindrical with short attenuate beak, usually deeply constricted 
toward center and flaring open when dried; seed 5 to 6 by 6 to 8 mm., thin, 
dull black, very narrow marginal wing. (Pis. 26, 27, and 29,A) 

Type : North end Fish Springs or House Range, western Utah, G. K. Gil- 
bert, August 1901, U. S. Natl. Herbarium No. 392927. 

Range: Burbank and Garrison, southwestern Millard County, north to 
Fish Springs, Juab County, Utah. The writer has studied the species in the 
southern end of the House Range, Millard County, Utah, at 4,700 to 6,200 
feet elevation. In this section, it is thinly scattered in the low chaparral, on 
gravelly bajadas, and mountain slopes. Although the foliage of the species 
is rather attractive, its greenish-yellow flowers detract considerably from its 
decorative value. The species usually flowers around the middle of June. 

Trelease's photograph (32, pi. 12) shows a potted plant of Yucca gilber- 
tiana that is similar to many of these plants in their native habitat and clearly 
depicts the characteristic rough and ragged appearance. From a short distance 
such plants have the vegetative appearance of Y. baccata. Y. gilbertiana has 
a raceme similar to that of Y. glauca. 

The writer has seen only one striking variant of Y. gilbertiana. This plant, 
occurring in the House Range, was well isolated from other species, and no 
plant in the vicinity exhibited evidence of intergrading into another species. 
The clump of the variant was very open, with scattered rosettes that were un- 
doubtedly of rhizomatous origin. The rosettes were comparatively few-leaved 
and the leaves extremely short. Although the racemes were borne above the 
foliage, this characteristic was undoubtedly caused by the short leaves, as the 
scape and inflorescence were of normal length. 

Comparison of Yucca gilbertiana, Yucca neomexicana, and 
Yucca barrimaniae 

In the range indicated above, Yucca gilbertiana was not found to be 
associated with any other yucca. Transitional forms between Y. gilbertiana 
and Y. neomexicana are commonly admixed with Y. baccata, Y. angustisjima, 
and Y. elata. No evidence of hybridization between any of the latter species 
and transitional forms of Y. gilbertiana-neomexicana has been observed. 

Yucca gilbertiana, Y. neomexicana, and Y. barrimaniae Trel. (31) 18 com- 
pose a confusing alliance, which has frequently been considered a single 
species, Y. harrimaniae. Field, nursery, and laboratory studies, however, 
indicate that Y. gilbertiana and Y. neomexicana are easily separated by specific 
characters, while Y. hatrimaniae appears to be an intermediate, variable race. 
The dumps of Y. gilbertiana are dense and small, whereas those of Y. neo- 



is Type: Helper, Carbon County, Utah, Trelease, Aug. 4, 1899, and Sept 7, 1901; Mo. 
Bot. Gard. Nos. 148566 to 148572 



46 AGR. MONOGRAPH 17, U. S. DEPT. OF AGRICULTURE 

mexicana are open and large ; the heads of leaves of Y. gilbertiana are mainly 
of aerial origin, but those of Y. neomexicana are usually of rhizomatous 
origin; the leaf blades of Y. gilbert iana are 18 to 43 mm. wide, lanceolate to 
spaculate-lanceolate, somewhat wavy, very glaucous, and rather light green 
(pi. 29,^1), but those of Y. neomexicana are 7 to 20 mm. wide, linear to 
narrowly lanceolate, straight, lightly glaucous, and dark green (pi. 29,-D, and 
29,£) ; the marginal leaf fibers of Y. gilbert iana are coarse and curly, whereas 
those of Y. neoniexicana are thin and straight; the racemes of Y. gilbertiana 
are densely flowered (average, 38 flowers) and extend from 10 to 20 cm. 
into the foliage, whereas the racemes of Y. neomexicana are thinly flowered 
(average, 27 flowers), mainly well above the foliage and never extend into 
the foliage; and the basic flower color of Y. gilbertiana is greenish yellow 
and that of Y. neomexicana is white. The filaments of Y. neomexicana are 
considerably more hirsute than those of Y. gilbertiana. 

Relative to Y. harrimaniae, Trelease (31, p. 59) described it as ". . . flower- 
ing from close to the base," whereas he (32) does not give the flowering 
habit of Y. gilbertiana. He reported the flowers of Y. harrimaniae as greenish, 
and the leaves as usually 6 to 15 or even 40 mm. wide; he does not note the 
flower color of Y . gilbertiana, but he reported the leaves of the latter as 20 mm. 
wide. With regard to Y. gilbertiana, Trelease (32, p. 225) seated it to be 

a rather striking form in its very rough papillate leaves, those of the type 
[harrimaniae'] being entirely smooth or with a few mostly low papillae near 
the apex." 

Rydberg (28, p. 170) raised fche status of Y. gilbertiana from a variety of Y. 
harrimaniae to that of a distinct species, and described its leaves as "linear, 
about 4.5 dm. long, 2 cm. wide, openly concave ..." He stated thai 
harrimaniae leaves are "linear to lanceolate, 6-40 mm. wide ..." 

McKelvey (25, p. 145) merged Y. gilbertiana and Y. neomexicana under 
Y. harrimaniae and reported, "... to a varying degree papillae are present on 
leaves of this species throughout its range." McKelvey also stated, "The 
flowers of this type [gilbertiana] must be considered similar to those of 
Y. Harrimaniae until proof is forthcoming that at anthesis the style is white 
rather than the apple-green color of the typical form." 

The writer found that the leaves of Y. gilbertiana were coarser and rougher 
than those of Y. neomexicana, but that the character depends upon the 
coarseness of striation, rather than papillae, which were present in varying 
degrees in both species, as McKelvey observed. The ovaries of Y. gilbertiana 
were pale green and the styles were slightly lighter green or occasionally white, 
whereas in Y. neomexicana both ovaries and styles were pale green or white. 

According to McKelvey (25) D. M. Andrews, Rockmont Nursery, Boulder, 
Colo., is of the opinion that Y . harrimaniae and Y. neomexicana are distinct 
species. In support of this contention, Mr. Andrews pointed out that the 
range of Y. neomexicana (New Mexico, Oklahoma, southeastern Colorado) 
is widely separated from that of Y. harrimaniae (west-central Colorado, Utah) 
and that the leaves of Y. harrimaniae are more spatulate than those of Y . 
neomexicana. Although there is no question that the leaves of the majority of 
plants in Utah and western Colorado are more spatulate than those of the 
eastern plants, many leaves from plants in the former sections are indistinguish- 
able in shape and size from those of the latter range. McKelvey cited spec- 
imens of this group of plants in Costilla and Huerfano Counties, Colo., that 
were approximately midway between the eastern and western ranges of the 
plants. 



YUCCAS OF THE SOUTHWEST 47 

Relative to Y. harrimaniae and the type locality, (Helper, Utah), McKelvey 
(25) reported that the plants begin 3 or 4 miles north of the town and that 
the largest ones produced procumbent stems, not more than a meler long. 
McKelvey further stated that in these plants flowering started well down in 
the foliage and that the styles were nearly oblong-cylindrical, very slender, 
and apple green (darker than the ovary at anthesis). In August 1948, the 
writer found abundant plants within the city limits, slightly north of the 
railroad yards. These plants were very similar to Y. neomexicana in rosette 
distribution (open clump), in leaf size and shape, in flowering stalk height, 
and in capsule size and shape. They differed only in that the leaves were 
slightly more filiferous and the fibers were somewhat curled. None of the 
plants had aerial stems, and the height of flowering could not be determined. 
Previous examinations of plants from the vicinity of Price, a nearby locality, 
showed that the racemes extended well into the foliage and that the styles were 
frequently quite tumid and varied in color from pale green to nearly white. 
It is the writer's opinion that the yuccas of Y. harrimaniae alliance in the 
environs of Helper and in other parts of Utah are hybrids or recombinations 
between Y. neomexicana and Y. gilbertiana, and that McKelvey's observations 
partly involved such plants. 

Apparent Hybrids of Yucca gilbertiana and Yucca neomexicana 

Apparent hybrids of Yucca neomexicana and Y. gilbertiana extend from the 
Duchesne and Uintah Counties, Utah, southwest to Iron, Piute, and to 
southwestern San Juan Counties, Utah. In the north and southeastern part 
of this range the plants closely approach Y. neomexicana, but in the central 
and southwest they are quite similar to Y. gilbertiana. The majority of plants 
resemble Y. gilbertiana in their small caespitose clump and long, broad leaves 
with coarse, curly fibers, and approach Y. neomexicana in their thinly flowered 
raceme and basically white flowers. Although the racemes usually extend 
slightly into the foliage, they rarely appear as low in Y. gilbertiana and not 
uncommonly it begins at foliage level or even several centime ers above the 
foliage. (Pis. 28, 29,£, and 29,C). Plants resembling Y. gilbertiana in all 
characters except height of flowering and flower color are particularly common 
from Manderfield south to Paragonah, Utah, whereas those differing from 
Y. neomexicana only in their dense caespitose clump, low flowering habit, and 
curly leaf marginal fibers are common in the vicinities of Price and Helper, 
and in San Juan County, Utah. 19 Rarely between Price and Helper are the 
clumps open as in Y. neomexicana. 

17. Yucca glauca 

(Great Plains yucca; locally, beargrass, soapweed) 

Yucca glauca Nutt., Fraser's Catalogue of New and Interesting Plants, No. 89. 
1813; reprinted in Pittonia 2: 115. 1890. 

Yucca angustifolia Pursh, Flora Amer. Sept., p. 227. 1814. 

Y. glauca Trel., Mo. Bot. Gard. Ann. Rpt. 13: 59 (in part). 1902. 

Plant acaulescent or with short stems (0.3 m.), solitary but soon clumped; 



19 Goosenecks, San Juan County, Utah, W. P. Cottam. April 30, 1935, Univ. Utah 
Herbarium, No. 5812. 



48 AGR. MONOGRAPH 17, U. S. DEPT. OF AGRICULTURE 

clump small (0.8 to 2.5 m. in diameter), dense (6 to 15 or more heads of 
leaves per square meter); head of leaves rather small; leaf blade 50 to 70 
(56.6) cm. long, 0.5 to 1.1 (0.9) cm. wide, linear, plano-convex, occasionally 
triquetrous or nearly flat, striate, divergently spreading, flexible, pale green 
or pallid; leaf margin white or greenish white, soon finely filiferous; flower- 
ing stalk 88 to 125 (107.6) cm. long; scape 24 to 53 (36.0) cm. long; 
inflorescence extending 5 to 21 (12.8) cm. into foliage, usually simple or 
rarely with few, abortive branchlets at base; flowers globose or campanulate, 
greenish white, commonly tinged with purple and shiny; perianth segments 
thick, brittle, broad, and acute; sepals 45 to 56 (50.8) mm. long, 26 to 33 
(28.5) mm. wide; petals 48 to 61 (53.9) mm. long, 31 to 42 (36.8) mm. 
wide; filaments 20 to 30 (21.1) mm. long; pistil 29 to 37 (33.1) mm. long; 
ovary 9 to 13 (10.8) mm. in diameter, obovoid, white or rarely greenish 
white, carpel sutures faint, anther depressions prominent, abruptly termin- 
ating in style; style 8 to 13 (10.2) mm. long, very tumid or swollen to 6 to 
10 (7.3) mm. in diameter toward center, dark green or rarely medium green; 
capsule 58 to 62 (60.1) mm. long, 45 to 53 (48.5) mm. in diameter, oblong- 
cylindrical, mucronate, with occasionally 1 or 2 valves slightly constricted; 
seed 7 to 9 by 8 to 10 mm., thin, smooth, dull black, broad marginal wing. 
(Pis. 30, 34,i4, 35,4 and 36,A). 

Type: 1,600 miles up Missouri River, about lat. 49, type specimen not 
located. McKelvey (25): ". . . most probably in the general region of Fort 
Mandan, North Dakota." 

Range: From Cottle and Floyd Counties, Tex., northwest through the 
Oklahoma Panhandle and extreme northeastern New Mexico, into the south- 
east quarter of Colorado. The species is also reported to occur in Kansas, 
Nebraska, northwestern Missouri, central South Dakota, and southern Wyo- 
ming. The writer has collected typical plants of the species on the eastern 
slopes of the Big Horn Mountains, north-central Wyoming, and in the eastern 
part of the Badlands National Monument, S. Dak. In the greater part of its 
range, Yucca glauca is rather thinly scattered on the rolling, well- drained 
grassland plains, at 1,600 to 2,800 feet elevation. In many areas, however, 
particularly in Colorado, it occurs in open woodlands at 2,000 to 6,500 feet 
elevation. Around Calhan, Simla, and Kiowa of El Paso and Elbert Counties, 
Colo., comparatively large, dense concentrations of Y. glauca are rather com- 
mon. Similar concentrations are reported in Logan, Phillips, and Sedgwick 
Counties. The species usually flowers between May 15 and June 30. 

On three occasions in the field and one in the nursery, the writer observed 
cases of phyllody in Y. glauca. In all instances the young inflorescence then 
was highly foliaceous. The inflorescence leaves were slightly shorter and 
broader than normal, and their apices are particularly pungent and curved 
abruptly inward, forming apical hooks. 

A striking variant patch of Y. glauca occurs near Kiowa, Colo. This varia- 
tion is described under Ecology and is probably caused by soil conditions. It 
mainly involves a change in plant habit and leaf character. 

Under Y. baileyi are described nanate plants that occur in Colorado at 
8,800 to 9,000 feet elevation. These may prove to be Y. glauca. 

Two forms of Y. glauca have been described. Variety stricta (Sims) 
Trelease (31), later renamed variety garneyi McKelvey (25), is mainly dis- 
tinguished by its robust growth and its conspicuous panicle with a long race- 
mose tip. It is reported to occur in Oklahoma, Kansas, New Mexico, and 
Colorado. Such plants (pi. 31) are rather uncommon, and the writer is of the 



YUCCAS OF THE SOUTHWEST 49 

opinion that they are results of hybridization or of particularly good growing 
conditions. They are usually found along roadsides or in areas where other 
vegetation is likewise exceptionally good. 

Throughout the greater part of northern New Mexico and extreme western 
Texas and to some extent into the Oklahoma Panhandle the plants usually 
identified as Y. glauca are extremely variable and lack constant characters. 
In general, the yuccas in these regions appear to be gradients of Y. glauca and 
closely allied species (see pis. 31-43). They are treated under transitional 
forms (p. 56). 

18. Yucca baileyi 
(Alpine yucca) 

Leaves linear, mainly long (29 to 56 cm.), narrow in proportion to length 

Y. BAILEYI. 
Leaves linear to oblanceolate, short (11 to 4l), broad in proportion to length.... 

Y. BAILEYI var. NAVAJOA. 

Yucca baileyi Woot. & Standi., U. S. Natl. Herbarium Contrib. 16: 114. 
1913. 

Yucca standleyi McKelvey, Yuccas Southwest. U. S. 2: 108. 1947. 

Plant acaulescent or rarely with short ascending stems, solitary but soon 
clumped; clump small (0.6 to 2.0 m. in diameter), dense (usually 3 to 18 
heads of leaves); leaf blade 20 to 56 (46.3) cm. long, 0.3 to 0.8 (0.6) cm. 
wide, linear, plano-convex, occasionally triquetrous or nearly flat, striate, di- 
vergently spreading, flexible, often falcate, pale green or yellow green; leaf 
margin white, soon finely filiferous; flowering stalk "35 to 120 (63) cm. long; 
scape 7 to 21 (14.3) cm. long; inflorescence a simple raceme extending 5 to 
20 (13.0) cm. into foliage, densely flowered; flowers campanulate or decidedly 
globose, greenish white, usually deeply tinged with purple, shiny; perianth 
segments rather thin, ovate or obovate; sepals 40 to 53 (46) mm. long, 18 
to 22 (20.1) mm. wide; petals 40 to 53 (44.1) mm. long, 26 to 31 (28.7) 
mm. wide; filaments 21 mm. long; pistil 27 to 29 (27.8) mm. long; ovary 7 to 
9 (8.1) mm. in diameter, stout, oblong or obovoid, very pale to fairly dark 
greenish white, carpel sutures well marked, anther depressions usually faint or 
rarely well marked, abruptly terminating (shouldered) in style; style 7 to 9 
(8.3) mm. long, slender, white or pale greenish white; capsule 44 to 70 (59) 
mm. long, 27 to 51 (38.5) mm. in diameter, oblong with oblate apex and 
mucronate tip, seldom constricted; seed 8 to 10 by 9 to 12 mm., thin, smooth, 
black with broad marginal wings. (Pis. 34,£, 35,5, $6,B, and 44.) 

Type: Dry slope in pine woods in the Tunitcha Mountains, San Juan 
County, N. Mex., P. C. Standley 7638, August 8, 1911, U. S. Natl. Herbarium 
No. 686602. 

Range: Coconino, Navajo, and Apache Counties, Ariz., east and northeast 
into McKinley, San Juan, and Valencia Counties, N. Mex., and La Plata, 
Archuleta, and San Juan Counties, Colo. It is also reported to occur in San 
Juan County, Utah, and to extend as far north, northeast, and east, respectively, 
as Mesa, Larimer, and Las Animas Counties, Colo. The species is thinly 
scattered in grasslands and woodlands at 4,000 to 8,000 or more feet elevation. 
It usually flowers from mid-May to mid-June. 

Although on several occasions, mainly in June 1945 and early August 
1948, the writer has studied the yuccas from Canon Gty, Fremont County, to 



50 AGR. MONOGRAPH 17, U. S. DEPT. OF AGRICULTURE 

Monarch Pass, Chaffee County, Colo., he has not found them in flower and 
has been unable to determine their identity. Yucca baileyi has been reported 
near Canon City (about 5,500 feet elevation), and the leaves, old flowering 
stalks, and capsules of the plants in the section appear to resemble those of 
that species. From Canon City west, however, the yuccas become increasingly 
small, until approximately 8 miles east of Monarch Pass they are quite dif- 
ferent from Y. baileyi. In the latter vicinity, the yuccas occur in small, dense 
patches on rocky slopes at 8,800 to 9.000 feet elevation. The plants form 
few rosettes, and the clumps are very open. The rosettes contain very few 
leaves; the leaf blades are short (25 to 33 (28.3) cm.) and very narrow 
(4 to 6 (5) mm.) but deeply concave; the flowering stalks are short 
(30 to 60 (40) cm.); and the capsules are small and commonly deeply con- 
stricted. In many respects the plants resemble Y . neomexicana. and, irrespective 
of high elevation and the gradual approach of the nanate condition, they may 
represent a transitional form of Y . bailey (or Y . glauca) and Y. neomexicana. 

Upon examination of the type specimen of Y. baileyi (U. S. Natl. Herbarium 
No. 686602), McKelvey (25) noted that the upper bracts were foliaceous 
and that the pistil was exceptionally large (5 cm. long) and similar in form 
to those found in the Baccatae. She further found that in the type locality 
(Tunitcha Mountains, N. Mex.) the plants rarely formed foliaceous bracts and 
none of the flowers had pistils similar in size and form to that of the type 
specimen. These conditions and the fact that Y. baccata was common in the 
type locality of Y . baileyi led McKelvey to the conclusion that the type col- 
lection represented a hybrid between Y . baccata and the capsular-fruited species 
of the area. She, therefore, concluded that the name Y. baileyi was inapplicable 
to the common capsular plant of the vicinity and named it Y. standleyi. 
McKelvey, however, noted that the original description of the type specimen 
of Y. baileyi was similar to that of Y. standleyi in plant and inflorescence 
habit and in leaf and fruit form and size. 

It is the writer's opinion that the Y. baileyi type differs from Y. standleyi 
only in its foliaceous bracts. Although he also finds the pistil of the Y . 
baileyi type specimen exceptionally large, he finds that its oblong ovary with 
very oblate apex and slender style is indistinguishable from that of other 
capsular yuccas of the type locality, rather than similar to or suggestive oi 
Y. baccata s fusiform ovary and style. It is very likely that the exceptionally 
large-sized pistil of the type specimen is caused by its age. In all yuccas, the 
perianth segments remain fresh several days following anthesis. During this 
period, presumably owing to fecundation, the ovary and, to some extent, the 
style develop very rapidly. Yucca flower measurements are usually made at 
anthesis at the height of the flowering season. 

McKelvey* s type specimen of Y. standleyi (No. 4609), Tunitcha Mountains, 
was collected May 24, 1934. Standley's type specimen of Y. baileyi was 
collected August 8, 1911. At such a late date, it is very likely that Standley 
had very little choice of flower material and was forced to take an old flower, 
whether or not the plant had foliaceous bracts. Except for the latter variation, 
which occasionally occurs in other yuccas, the Y. baileyi type specimen is similar 
to all other capsular-fruited plants in the type locality. It therefore appears 
unwarranted to rename the group in favor of the majority of plants, and 
exclude those exhibiting the minor variation. 

Yucca baileyi and Y. glauca are similar, and it would undoubtedly be difficult 
to distinguish hybrids between them or to determine which one of them 



YUCCAS OF THE SOUTHWEST 51 

was involved in hybrids with other species. In vicinities where Y. baileyi and 
Y . angustissima are relatively close together, however, the clumps of Y. 
baileyi are rather large and open, suggesting the influence of Y. angustissima. 
Such possible hybrids are discussed on page 56. 

Yucca baileyi variety navajoa 

Yucca baileyi var. navajoa. (Webber) comb. nov. 

Yucca navajoa Webber, Madrono 8: 105. 1945. 
Variety navajoa differs from typical Yucca baileyi in the following respects: 
(1) It is usually subcaulescent rather than acaulescent; (2) its dense clump 
is mainly caused by the branching of aerial stems (0.4 to 1.2 m. tall) rather 
than to branching of subterranean stems; (3) its heads of leaves are smaller 
and more symmetrical than those of the species; and (4) its leaves are broader 
(average, 8 mm.) in comparison to their length (average, 23 cm.) (frequently 
oblanceolate) than those of the species. The seeds of a single plant at the type 
locality were small (average, 6 by 7 mm.) and lacked the broad marginal wing 
characteristics of the species. (PI. 45.) 

Type: Five miles north of Tohatchi, McKinley County, N. Mex., Webber. 
1944; U. S. Natl. Herbarium No. 1872608. 

Range: Three miles north of Tohatchi, McKinley County, N. Mex., extend- 
ing north for approximately 6 miles and west into the Chuska Mountains, 
possibly as far as the Little Colorado Gorge, Coconino County, Ariz. The 
variety is usually scattered in chaparral and juniper woodlands on coarse 
gravelly soil or sandstone ledges, at 5,200 to 6,500 feet elevation. It usually 
flowers in early June. 

Although the leaves of variety navajoa are small, they are quite different 
from the small leaves of the dwarf forms of Y. baileyi, which occur in the 
vicinity of Keams Canyon and south of Holbrook, Ariz. The leaves of the 
dwarf plants are narrow and frequently needlelike, whereas those of variety 
navajoa are broad and commonly oblanceolate. Furthermore, variety navajoa 
always appears vigorous, having as large an inflorescence as and forming more 
rosettes than Y . baileyi ; the nanate forms of Y. baileyi appear stunted, having 
short inflorescences and few rosettes. 

In the vicinity of the Little Colorado Gorge, northwest of Cameron, 
Coconino County, Ariz., the yuccas resemble variety navajoa, but in many 
characters they appear midway between variety navajoa and Y. baileyi. These 
plants frequently have aerial stems 0.30 to 0.45 m. tall that are moderately 
branched. The leaves are short, narrow to broad, and linear to oblanceolate. 
These plants, however, do not have so many stems or rosettes as variety 
navajoa, and they do not form large, dense clumps. It appears very likely 
that these yuccas are border plants of the variety navajoa and that the range 
of the variety extends from the vicinity of Tohatchi, N. Mex., west through 
the Chuska Mountains to the Little Colorado Gorge. Kearney and Peebles 
(20) referred to the Little Colorado Gorge plants as an unnamed variety of 
Y. baileyi and described them as having exceptionally short and rather coarsely 
flliferous leaves. 20 McKelvey (25) believed they were juvenile forms of the 
species (Y. standleyi). 



20 Such coarse-fibered plants as Kearney and Peebles No. 12819, Sacaton Herbarium, 
U. S. Cotton Field Station, Sacaton, Ariz., arc quite rare in the locality. The leaves of the 
specimen are narrowly oblanceolate. 



52 AGR. MONOGRAPH 17, U. S. DEPT. OF AGRICULTURE 

The writer has found one striking variant in variety navajoa that differs 
from the normal in the same manner as the variant form of Y. gilbertiana 
described on page 45. This variant occurred in the center of a rather dense 
concentration of navajoa, approximately 5 miles north of Tohatchi, N. Mex. 
The plant had scattered rosettes of rhizomatous origin and lacked aerial 
stems. Its rosettes were small, and its leaves were extremely short and broad 
for their length. The scape and raceme were normal length and the raceme, 
probably owing to the short leaves, were borne at foliage level. 

19. Yucca angustissima 

Yucca angustissima Engelm. ex Trel., Mo. Bot. Gard. Ann. Rpt. 13: 58. 1902. 

Plant acaulescent or rarely with short (0.1 to 0.5 m.), procumbent stems; 
solitary but at length clumped; clump large, open with 20 or more heads of 
leaves at a density of 1.8 to 5.3 per square meter; head of leaves symmetrical; 
leaf blade 28 to 57 (44.8) cm. long, 0.4 to 0.8 (0.6) cm. wide, striate, plano- 
convex, rarely triquetrous or nearly flat, linear, rigidly divergent, flexible, pale 
blue green; leaf margin white or greenish white, eventually filiferous, the fibers 
fine, loosely curled; flowering stalk 90 to 122 (105) cm. long; scape 28 to 34 
(31.2) cm. long, slender; inflorescence simple or rarely with 2 or 3 short 
abortive branchlets toward base, 15 to 26 (20.3) cm. above foliage; flowers 
campanulate to globose, greenish white, the Outside tinged with purple and 
glossy; perianth segments elliptical, brittle; sepals 31 to 40 (37) mm. long, 
19 to 23 (22) mm. wide; petals 31 to 39 (35.1) mm. long, 28 to 32 (30.5) 
mm. wide; filaments 22 to 24 (23.1) mm. long; pistil 25 to 28 (27) mm. 
long; ovary 7.0 mm. in diameter, slender, ovoid, more or less tapering into 
style, pale green or nearly white; staminal depressions and carpel sutures 
moderately deep; style 4 to 6 (5) mm. long, slender or somewhat ovoid at 
union with ovary, nearly white or pale green; capsule 38 to 42 (41.3) mm. 
long, 30 to 33 (31) mm. in diameter, commonly with deep central constric- 
tion, dehiscent for entire length but twisted and flaring open above constric- 
tion; seed 5 to 7 by 7 to 8 mm., thin, dull black, with fairly narrow marginal 
wing. (Pis. 46, 34,C, and 35,C.) 

Type: Deserts of the Colorado River area, /. M. Biglow, 1854, Mo. Bot. 
Gard. No. 148375; McKelvey (25) near Picacho, Ariz. 

Range: Coconino, Mohave, and Yavapai Counties, Ariz., and in Wash- 
ington and Grand Counties, Utah. It is also reported to occur in Apache 
County, Ariz., and in Rio Arriba and Valencia Counties, N. Mex. The writer 
has collected specimens closely resembling the species in Navajo County, 
Ariz., and in southeastern Nevada. The plants in the latter areas, however, 
as well as those in New Mexico, Utah, and Apache County, Ariz., are quite 
variable, and the majority of them exhibit characters of Yucca glauca (or 
Y. baileyi) and Y. elata. The species is thinly scattered in grass, woodlands, 
and chaparral at 2,500 to 7,500 feet elevation. It usually flowers in May or 
early June. 

With reference to the habit of Y. angustissima, McKelvey (25, p. 106) 
stated "Individual plants as a rule produce a single head of leaves or but 
few heads in one clump." Although plants with a single head of leaves 
occur, those with many heads, widely scattered but connected by subterranean 
stems, are far more abundant. In the majority of yuccas, the area of the 



YUCCAS OF THE SOUTHWEST 53 

clump (or a single plant) is easily determined by tracing minor variations 
of the individual. 

In areas where Y. angustissima is removed from other capsular forms of 
yucca, it exhibits very little variation. Possibly the greatest variations in the 
species are in the color of its style, the amount of capsular constriction, and the 
branching of the inflorescence. The styles vary from pure white to fairly 
dark green, and they are occasionally rather yellowish. All plants have con- 
stricted capsules, but the degree of constriction varies from shallow to very 
deep, and the number of capsules constricted varies from approximately 60 to 
100 percent. In several localities, mainly around Peach Springs, Seligman, 
and Paul den, Ariz., the inflorescences are unbranched, and elsewhere scattered 
plants with inflorescences having a few short branchlets at or near the base 
occasionally occur. 

From Whiteriver north to McNary and Show Low, Ariz., the yuccas are 
very confusing. The majority of plants in this area occur between 6,000 and 
6,500 feet elevation in open woodlands of deep, coarse sand, which is often 
packed solid, like hardpan. The yuccas are either typical Y. angustissima or 
plants that exhibit characters of both Y. angustissima and Y. elata. Both types 
of plants have small heads of leaves, and the leaves are short and narrow. 
Plants of Y. angustissima have typical flowers and capsules, but their racemes 
may be either normal size or extremely short. Except for their slender ovoid 
ovaries that taper into the styles and for their small constricted capsules, both 
types are quite similar to the dwarf plants of Y. baileyi occurring a few miles 
north, between Snowflake and Holbrook. 

Yucca angustissima is similar to Y. glauca and Y. baileyi, and it is the 
writer's opinion that much of the difficulty in separating the species is owing 
to hybridization between members of the three species (see pi. 36,D, E, and F) . 
Apparent hybrids and transitional forms are described on page 56. Y. angus- 
tissima is occasionally admixed with or in close proximity to Y. elata, and 
these two species also intergrade (see pi. 37,E, and F) p. 62). 

20. Yucca constricta 

Yucca constricta Buckl., Phila. Acad. Nat. Sci. Proc, 1862: 8. 1863- 

Yucca constricta Trel., Mo. Bot. Gard. Ann. Rept. 13: 54-56 (in part). 
1902. 

Plant acaulescent or with short procumbent stems 0.2 to 0.4 m. long; dump 
small to large, open with 1 to 20 heads of leaves 0.2 to 0.9 m. apart; head of 
leaves large, asymmetrical or rarely small and symmetrical; leaf blade 29 to 65 
(44.1) cm. long, 0.8 to 1.5 (1.1) cm. wide, linear or somewhat broader toward 
center, nearly flat or plano-convex, occasionally triquetrous, striate. 
rigidly divergent, flexible, pale to dark green; leaf margin white or green, 
eventually filiferous, the fiber fine, straight or curled, soon eroding away; 
flowering stalk 1.53 to 3.20 (2.07) m. long, slender and weak; scape 0.95 to 
2.00 (1.31) m. long; inflorescence paniculate; panicle 0.20 to 0.46 m. above 
foliage, 0.45 to 1.20 (0.76) m. long, occasionally 0.7 m. in diameter at base, 
ovoid or broadly conoid, open-flowered, with few (8 to 15) branchlets; flowers 
small, hemispherical or cup-shaped, pale greenish white; perianth segments 
thin, acute, ellipsoidal, sepals 35 to-45 (41.5) mm. long, 11 to 21 (16.1) mm 
wide; petals 37 to 48 (42.4) mm. long, 17 to 26 (22.0) mm. wide; filaments 



54 AGR. MONOGRAPH 17, U. S. DEPT. OF AGRICULTURE 

17 to 22 (19.1) mm. long; pistil 25 to 38 (27.1) mm. long; ovary 5 to 6 
(5.5) mm. in diameter, slender, cylindrical with oblate apices, pale green, 
carpel sutures prominent, anther depressions faint or wanting, abruptly 
terminating (shouldered) in style; style 8 to 11 (9.1) mm. long, cylindrical 
with basal lobes extending slightly over ovary, whitish green or pale green 
(darker than ovary); capsule 46 to 63 (53) mm. long, 36 to 43 (39.5) mm. 
in diameter, oblong-cylindrical, mucronate, 17 to 92 percent constricted; seed 
7 to 9 by 9 to 15 mm., thin, glossy black, with broad marginal wing. (Pis. 
34,1), 35,D, and 36, C, and 47.) 

Type: Western Texas, S. B. Buckley, 1861-62, Elias Durand Herbarium. 
(McKelvey (25) thought it was from the eastern end of Edwards Plateau.) 

Range: Gillespie County south to southern Dimmit County and southwest 
to the vicinity of Del Rio, Val Verde County, Tex. The species is also reported 
to extend from the latter section as far east as Galveston County; as far 
west as the Pecos River basin; and as far north as the Oklahoma-Texas State 
line. In the two latter sections, however, the species is extremely variable 
and blends into Yucca glauca and Y. data. The species is thinly scattered in 
brushwood and grasslands at 900 to 2,400 feet elevation. The plants usually 
flower in May. 

McKelvey (25, p. 130) stated, "The leaves of the type sheet of Y. con- 
stricta are grass-like and, therefore, do not conform to those attributed to that 
species by Trelease which (in his description) are 'rather rigidly divergent,' 
(in his key) 'more rigid and spreading' than the 'grass-like' ones of his Y. 
tenuistyla. Trelease's concept (1902) of Buckley's species and the identifica- 
tion of Buckley's plant by Engelmann (1873), Baker (1880), Sargent (1896) 
and Trelease (1898) with the one now called Y. data, certainly indicates 
that Buckley's type was not examined by any of these botanists; for the foliage 
of Y. elata is wiry and, where the blade is least thickened, flexible at most; 
it is never grass-like in the sense used by Trelease and the present writer; 
Buckley only refers to the length and breadth of the leaves, not to their 
texture." 

The writer has indicated that he has been unable to separate many species 
by their leaf texture. Among these plants are such species as Y. glauca, Y. 
baileyi, Y. angustissima, Y. elata, and Y. constricta. The texture, as well as 
the habit, size, and shape of the leaves in these species, is extremely variable. 
In all of them the leaves can best be described as rigidly divergent, with 
common variations being more or less rigid and spreading, wiry or grasslike. 
In general the leaves of nanate forms are very rigid, firm, and divergent, 
whereas those of normal-sized plants are divergent but flexible. The leaves of 
exceptionally longleaved plants are either rigid, flexible, and divergent or 
wiry, flexible, and drooping. Juvenile and, frequently, mature leaves, especially 
in heads where the leaves are twisted, appear grasslike. Aside from juvenile 
plants, the only truly flaccid or grasslike-leaved plants the writer has found in 
these species are variants of Y. glauca occurring near Kiowa, Colo. Undoubtedly 
the variation is caused by environmental conditions. It is discussed under the 
section entitled "Ecology." 

McKelvey (25 ) also described the leaves of Y. arkansana Trel. as grasslike 
and those of Y. hanimaniae, Y . neomexicana, and Y. gilbertiana as somewhat 
intermediate between grasslike and wiry. Y. arkansana occurs somewhat east 
of the range covered by the present monograph, and the writer has not studied 
the species. Many plants of Y. neomexicana, especially those with fine fibers 



YUCCAS OF THE SOUTHWEST 55 

that occur in New Mexico, Oklahoma, and eastern Colorado, are somewhat 
grasslike. The plants occurring in Utah, however, especially those with coarse 
fibers (Y. gilbertiana) , are rather rigid or frequently very rigid. 

The writer has not seen the type specimen of Y. con strict a. If the leaves ot 
the type are strictly grasslike, then he is convinced that few, very probably 
no, similar plants occur in the territory he has covered (west of Gillespie 
County, south to southern Dimmit County, Tex.). If in such species as Y. elata, 
Y . glauca, Y . angustissima, Y. constricta. and even Y. neomexicana, minor leaf 
characters (known to be variable) must conform to those of the type, then the 
majority of plants now identified as these species must be excluded. Buckley 
in his description of Y. constricts does not refer to leaf texture. Noting the 
similarity of Y. tenuistyla Trel. to Y. con strict a and of Y . louisianensis Trel. to 
Y. tenuistyla, McKelvey (25) concluded that Y. tenuistyla and Y. louisianensis 
were, possible synonyms of Y. constricta. Y . tenuistyla and Y. louisianensis 
occur east of the region surveyed by the writer, and, therefore, they have not 
been included in this study. 

Yucca glauca Alliance 
Species 

In the writer's opinion, Yucca angustissima, Y. bailey i. Y. glauca, and Y . 
constricta form an alliance of closely related species that are quite similar in 
many characters and frequently intermingle with each other. Y. elata is closely 
related to the alliance and possibly should be included in it. It freely inter- 
mingles with all members of the alliance and is quite similar morphologically. 
McKelvey (25) divided the Chaenocarpa section into six series, based mainly 
upon leaf and style differentiation. Y. angustissima and Y. baileyi (Y. standi 'eyi 
McKelvey) were placed in the Elatae series, distinguishd by wiry leaves and 
white to pale-green, rather slender styles and closely related to Y. elata and 
several new species. Y. constricta was placed in a series of its own, Constrictae, 
distinguished by slender, white or greenish styles and grasslike foliage and 
rather closely related to Y. angustissima, Y. standleyi, and Y. barrimaniae. Y. 
glauca was placed in Glaucae series, distinguished by stout, ovoid, apple-green 
styles and wiry leaves. These conclusions differ from those of the writer 
mainly in that Y. angustissima and Y. baileyi are more closely related to Y. 
elata than to Y. constricta and Y. glauca and that Y. glauca is quite distinct, 
being separated by Y. barrimaniae and several new intermediate species. 

Typical plants of the Y. glauca alliance may easily be distinguished by the 
habit in connection with other characters. Y. glauca forms a small dense clump, 
has a racemose inflorescence extending well into the foliage, a broad-shoul- 
dered ovary, and very dark green tumid style. Y. baileyi forms an equally dense 
clump, with a raceme extending into the foliage, a broad-shouldered ovary, 
somewhat smaller leaves, shorter flowering stalk, and pale-green or white, 
comparatively slender style. Y. angustissima forms an open clump, has a 
racemose inflorescence borne above the foliage, slender ovary tapering into 
the style, pale-colored, rarely swollen style, and smaller, usually constricted 
capsules. Y. constricta forms an open clump, has a panicled inflorescence with 
exceptionally long peduncle, slender ovary, pale-colored, slender style, and 
smaller constricted capsules. 



56 AGR. MONOGRAPH 17, U. S. DEPT. OF AGRICULTURE 

Apparent Hybrids 

Yucca glauca Trel., Mo. Bot. Gard. Ann. Rpt. 13: 59 (in part). 1902. 
Y. intermedia McKelvey, Yuccas Southwest. U. S. 2: 116. 1947. 

Throughout the northern half of New Mexico and well into adjacent parts 
of Texas and Oklahoma and in the southwestern corner of Utah and adjacent 
Nevada, the majority of capsular-fruited yuccas lack constant characters and 
they are difficult to identify or classify as named species or forms. Although 
occasionally a plant or a group of plants approaches a named type, the majority 
of them appear to be hybrids or hybrid derivatives. Many of the apparent 
hybrids are midway between two named types, whereas others appear to be 
of quite remote hybrid ancestry, and still others to have involved several 
named types in their origin. The large area in New Mexico and adjacent 
Texas appears to be mainly a transitional zone between Yucca glauca, Y. 
baileyi, and Y. angustissima, and to a lesser degree between these members of 
the Y. glauca alliance and Y. elata. The area in southwestern Utah and 
adjacent Nevada appears to be a remote section of Y. elata, in which hybrids 
and recombinations of Y. angustissima and Y. elata are plentiful. The rel- 
atively large area between Lamesa and Abilene, Tex., and south to the 
vicinities of McCamey and Junction, Tex., appears to be a zone of hybrids 
between Y. glauca and Y. constricta. The fourth transitional zone appears 
to be between members of the Y. glauca alliance and Y. elata. This zone 
varies between 25 to 100 miles wide, along an imaginary line from Prescott, 
Ariz., to Roswell, N. Mex., and east into Texas. 

On the United States Department of Agriculture Field Station grounds at 
Tucumcari, N. Mex., is a patch of native yuccas that clearly illustrates the 
perplexity of the majority of yuccas in the transitional zones. This patch 
covers approximately 30 to 35 acres of gentle sloping grassland, at about 4,100 
feet elevation. Throughout the field, the soil appears to be fairly deep, 
sandy loam. The common and outstanding variations in this field are as 
follows: Leaf blade 25 to 64 cm. long, 0.2 to 1.3 cm. wide; flowering stalks 
30 to 132 cm. long; scape to 79 cm. long; panicle 30 to 85 cm. long; in- 
florescence 35 cm. within foliage to 38 cm. above; perianth segments 24 to 

57 mm. long and 14 to 33 mm. wide; filaments 15 to 26 mm. long, pistil 
20 to 34 mm. long; ovary 6 to 10 mm. in diameter; style 5 to 12 mm. long, 
2 to 6 mm. in diameter; capsule 45 to 75 mm. long, 36 to 50 mm. in diameter, 
constricted or not constricted and widely dehiscent or completely indehiscent. 

The majority of plants resemble Y. glauca, with moderately branched pan- 
icles, although occasionally plants resemble Y. angustissima, Y, baileyi, and 
Y. constricta. Each form exhibits many characters that intergrade into the 
other form, and only a few plants approach typical species. All of the plants 
resembling Y. glauca differ from typical Y. glauca at least in their larger, 
more or less open clump and their pale-green or white, moderately tumid or 
slender styles. Plants resembling Y. angustissima differ from typical Y. 
angustissima mainly in their larger, nonconstricted capsules, whereas a few 
differ in their greener and more tumid styles. Plants that resemble Y. glauca 
and Y. baileyi are frequently very similar and are mainly distinguishable by 
the slender, white style and broad, cylindrical capsules of Y. baileyi. Plants 
similar to Y. constricta have long peduncles and fairly broad panicles. In 
many respects these plants appear intermediate between Y. glauca and Y. 
constricta. (Pis. 32, 33, 34,E, 35,E, and 36,D.) 



YUCCAS OF THE SOUTHWEST 57 

Throughout the transitional zone, definable forms occur. These forms 
may occupy a relatively large area or be limited to comparatively small patches. 
In the northwest corner of New Mexico, all plants belong to a form that 
the writer refers to as Aztec type. This type differs from described members 
of the Y. glauca alliance in that its leaves, inflorescences, and flowers are 
shorter and smaller than other members of the group. It is similar to Y. 
angustissima in its large, open clump and constricted capsules. It resembles 
Y. glauca and Y. baileyi variety navajoa in that its flowers extend well into 
the leaves. A second form, referred to as the Albuquerque type, occupies a 
narrow zone along the Rio Grande south of Albuquerque. These plants differ 
from those in the eastern part of New Mexico mainly in their smaller, denser 
clump; their white rather than green basic flower color, and their smaller, 
mainly constricted capsules. The type is more variable than the Aztec type. 
In the vicinities of Nara Visa and Pasamonte, N. Mex., large areas of late- 
flowering plants are common. These plants not only flower 2 or 3 weeks 
later than the majority of plants in northeastern New Mexico, but require 
approximately 2 more weeks to complete flowering. They have larger and 
more branched inflorescences than other members of the Y. glauca alliance, 
and possibly represent a third form. Other possible forms, such as sub- 
caulescent, short or narrowleaved types, are scattered throughout the transi- 
tional zones, but these forms are more variable and do not occur in concentra- 
tions of appreciable size. 

Relative to Y. glauca, Trelease (31, p. 60) stated that the inflorescence is 
"... almost invariably marked by a simple inflorescence, not carried on a 
scape above the cluster of leaves." He described the capsules as large, oblong, 
and usually not constricted. Trelease's illustration (his pi. 24, 1 and 2) of 
capsules of Y. angustissima from near Grand Canyon, Ariz., and of Y. glauca 
from Manitou, Colo., are typical of the types. His plate of Y. glauca (pi. 
23, 2) near Albuquerque, N. Mex., appears to have mainly constricted cap- 
sules. With regard to Y. angustissima, Trelease (31, p. 58) stated, "In habit, 
this species . . . recalls the narrow-leaved form of Y. glauca as found, for 
example, about Albuquerque, N. Mex., ... it differs in its more frequently 
branched inflorescence, oblong (white?) style, and smaller capsule and seed 
..." The plants in the vicinity of Albuquerque are extremely variable and 
more or less midway between Y. glauca and Y. angustissima. Although plants 
with simple inflorescences predominate, those with branched ones are plenti- 
ful, and it would be difficult to find typical Y. glauca capsules in the vicinity. 
Furthermore, no plants in the vicinity of Albuquerque have the tumid, dark- 
green style characteristic of Y. glauca. 

With respect to Y. angustissima and Y. baileyi Benson and Darrow (3, p. 
66) stated, "... they may be merely transitional forms between this [glauca'] 
and Yucca elata . . . Intermediate types between these forms and Yucca data 
are found along the southern limits of the range in Arizona." Since the char 
acteristics of these intermediate types are quite distinct from the differences 
between Y. glauca and other members of the Y. glauca alliance, the writer 
does not believe that Y. elata was involved in the origin of any member of the 
Y. glauca alliance. 

Between the range of Y. glauca and the ranges of Y. angustissima and Y. 
baileyi McKelvey (25, p. 117) separated Y. intermedia. She characterized her 
species thus: "... inflorescence is a tall, vigorous, simple raceme but a raceme 
with one to several short, basal branchlets is not unusual and a true panicle is 



58 AGR. MONOGRAPH 17, U. S. DEFT. OF AGRICULTURE 

occasional throughout the plant's range; the inflorescence proper starts among 
or immediately above the foliage; the plant forms small to medium size 
clumps with but few heads of leaves." According to McKelvey, the species 
differed from Y. glauca in its more tidy appearance and in its cylindric, white 
or pale-green style; from Y. baileyi (Y. standleyi McKelvey) in its less rigid, 
less filiferous leaves, which do not approach triquetrous; in its head of leaves 
(V. standleyi, hemispherical) that is constricted below and spreading above; 
and in its considerably longer more branched inflorescence; and from Y. 
angustissima in its larger, less constricted capsules, which have stronger, 
thicker walls, darker rougher epidermis, etc.; in its longer inflorescence (1.5 
m. as against 1.3 m. or usually less), which has a greater tendency to branch; 
and in its slightly longer and somewhat heavier pistil. 

In central New Mexico, McKelvey {25) separated Y. intermedia variety 
ramosa from the species by greater vigor of plant, foliage, and inflorescence. 
In characterizing the variety McKelvey (p. 120) stated, "Inflorescence on young 
plants sometimes racemose or again with a few basal branchlets, on large, 
mature specimens usually much branched from base upward, ..." and, "... 
the inflorescence is, for the most part, a large, broad panicle, branched for the 
greater part of its length, but from this form we also note occasional grada- 
tions to the simple but vigorous raceme." 

In all yuccas the size and shape of the leaf head is largely determined by 
its age and the amount of rainfall. The inflorescences of young plants are 
usually more branched and vigorous than those of matured plants. None of 
the characters separating Y. intermedia are characteristic of only the species. 
They appear to be combinations of the characters of members of the Y. glauca 
alliance. All characteristics of variety ramosa, especially that of McKelvey's 
description of the inflorescence, strongly suggest an intermingling of Y. 
elata and members of the Y. glauca alliance. Transitional plants similar to 
Y. intermedia are shown in plates 38, 39, and 48, and plants similar to 
variety ramosa are discussed under Y. elata and shown in plates 40 and 41. 

The panicles of Y. constricta are pyramidal or short and broadly conoidal 
in shape. They usually have between 8 to 15 branches, which vary from 10 
to 40 cm. in length. The inflorescence of Y. glauca is mainly racemose, or at 
the most has 4 short, abortive branchlets toward the base. The flowers and 
capsules of Y. glauca are considerably larger than those of Y. constricta, and 
the capsules of Y. glauca are seldom constricted, whereas those of Y. constricta 
are commonly deeply constricted. Within the area between McCamey, Junction, 
Lamesa, and Abilene, Tex., plants gradually change from Y. glauca in the 
north to Y. constricta toward the southeast. 

This transition mainly involves the change of large, open clump, slender, 
weak, tall inflorescence, panicle, and long scape, characteristic of Y. constricta, 
at Uvalde, to small, dense clump, stout short inflorescence, raceme, and short 
scape of Y. glauca at Hadley, Tex. In the center of this area, between Runnels 
and Upton Counties, the majority of plants are midway between Y. glauca and 
Y. constricta, but plants resembling either species in one or more characters 
are not uncommon. Plants with racemose and paniculate inflorescences occur 
in approximately equal numbers. The panicles are raggedly pyramidal in 
shape and have from 8 to 15 branches that are 10 to 35 cm. long. The majority 
of flowers and capsules appear to be intermediate in size, and the capsules 
are only moderately constricted. The styles are pale green or white, slender 
or somewhat swollen toward the center. North of Runnels and Upton 



YUCCAS OF THE SOUTHWEST 59 

Counties, the plants become more like Y. glauca, but scattered intermediates 
or plants resembling Y . constricta in one or more characters spread well into 
the western part of the Y. glauca alliance transitional zone. In New Mexico 
plants that resemble intermediates are fairly common in the vicinities of Lingo 
and Blui.t, and they extend as far north as Tucumcari and Mosquero. South 
of Runnels and Upton Counties, the plants rapidly approach Y. co?istricta. 
However, variable plants, possibly showing the influence of either Y. glauca 
or Y. elata, are found in the vicinities of Ozona and Del Rio, Tex. These 
plants have somewhat shorter flowering stalks (averages of 1.1 and 1.8 m., re- 
spectively), shorter scapes (0.8 and 0.95 m.), and larger and less constricted 
capsules. In the northwestern and somewhat in the southwestern part of the 
zone, transitional forms involving Y. elata are rather common. 

McKelvey (25) cited several apparent hybrids between Y. constricta and 
Y. rup'icola. Under Y . reverchoni, possible Y. rupicola X Y> constricta-glauca 
hybrids are discussed. These plants mainly resembled (1) rupicola, but had 
small, greenish, globose flowers, with short styles like Y. glauca-Y. constricta: 
or (2) Y. glauca, but had nonfiliferous, frequently somewhat denticulate leaves 
similar to those of Y. rupicola. 

In addition to the foregoing associations, members of the Y. glauca alliance 
are occasionally admixed with the capsular- fruited species Y. neomexicana and 
Y. elata and the fleshy-fruited species Y. torreyi, Y. baccata, and Y. treculeana. 
Possible hybrids between Y. elata and members of the Y. glauca alliance are 
discussed under Y. elata. No evidence of hybrids between capsular- and fleshy- 
fruited yuccas has been found. 

21. Yucca elata 

(Palmella, soapweed, or soaptree) 

Yucca elata Engelm., Bot. Gaz. 7: 17. 1882. 

Yucca angustifolia var. radiosa Engelm., in King, U. S. Geol. Expl. 40th 
Par. Rpt. 5: 496. 1871. 

Y. angustifolia var. elata Engelm., St. Louis Acad. Sci. Trans. 3: 50. 1878. 
Y. radiosa (Engelm.) Trel., Mo. Bot. Gard. Ann. Rpt. 3: 163. 1892; 
13: 56. 1902. 

Plant caulescent or arborescent, solitary but soon clumped; clump small (0.5 
to 2.5 m., or rarely up to 4.0 m. in diameter), fairly open (0.3 to 1.2 heads of 
leaves and stems per square meter); stems mainly 0.9 to 4.5 m. tall, trunklike, 
erect, with 1 to 5 short, assurgent branches toward the top, or rarely un- 
branched and up to 7.8 m. tall; head of leaves large, symmetrical; leaf blade 
commonly 30 to 95 cm. long, 0.4 to 2.5 cm. wide, linear, plano-convex or 
piano-keeled, striate, rigidly divergent, flexible, pale green; leaf margin white 
or greenish white, soon finely filiferous; flowering stalk 1.0 to 5.4 (2.9) m. 
tall, scape 0.9 to 3.0 (1.4) m. long, stout, mainly 2.5 to 6.5 cm. in diameter, 
rarely up to 12.0 cm. in diameter, greatly exceeding the foliage; panicle 0.5 
to 3.5 (1.3) m. long, 0.3 to 0.6 m. in diameter, with 20 to 45 branchlets, 
ellipsoidal or broadest toward center; flowers many, campanulate or somewhat 
globose, white or cream, rarely greenish or tinged with pink, perianth segments 
narrow-elliptical to broad-elliptical or even ovate, acute; sepals 32 to 57 (43-8) 
mm. long, 14 to 27 (19.6) mm. wide, petals 34 to 57 (44.8) mm. long, 20 



60 AGR. MONOGRAPH 17, U. S, DEPT, OF AGRICULTURE 

to 35 (27.8) mm. wide; filaments 16 to 30 (20.6) mm. long, slender; pistil 
22 to 33 (26.8) mm. long; ovary 6 to 10 (8.2) mm. in diameter, slender, 
white or pale green, oblong-cylindrical, abruptly terminating in style, with deep 
carpel sutures and anther depressions prominent or wanting; style 6 to 11 
(8.8) mm. long, shape variable, slender and oblong-cylindric or. stoutish, 
white or pale green; capsule 40 to 70 (52) mm. long, 35 to 58 (47) mm. in 
diameter, oblong-cylindric, mucronate, commonly symmetrical, rarely con- 
stricted; seed 7 to 10 by 9 to 14 mm., thin, dull black, with broad marginal 
wing. (Pis. 37, A-C, 49, and 50.) 

Type: Extending from west Texas to Utah, Arizona, and northern Mexico, 
/. T. Rotbrock 382, 1874, U. S. Natl. Herbarium No. 35891. Lectotype: Mc- 
Kelvey, Camp Grant, Graham County, Ariz. 

Range: Within the United States, Yucca elata has two well-separated ranges. 
The larger range forms a wide belt extending from Pecos County, Tex., north- 
west and west through southern and south-central New Mexico to north-central 
Yuma County and south-central Maricopa County, Ariz. The second range is a 
relatively small one, occurring in western Washington and Iron Counties, Utah. 
Although the species is usually scattered, it frequently forms dense concentra- 
tions covering many square miles. Within the large southern belt Y. elata 
occurs on gently sloping deserts and desert grasslands, at 1,500 to 6,000 feet 
elevation, and usually flowers between the first of June and the middle of July. 

Yucca elata is a well-defined species that may easily be recognized by its 
(1) well-developed trunklike stem; (2) small, open clump of few heads of 
leaves; (3) long, flexible leaves, with filiferous margins, (4) tall flowering 
stalk with large, ellipsoidal panicle; and (5) capsular fruit. However, in the 
northern limits of its range, plants of the species frequently appear weak and 
degenerate. Such plants are particularly common in New Mexico, especially 
between Picacho, Lincoln County, and Elkins, Chaves County, and between 
Ancho, Lincoln County, and Claunch, Socorro County. These plants rarely 
form a clump, and their stems are slender, frequently leaning or procumbent, 
and rarely over a meter long. The heads of leaves are small, with but a few 
short leaves. On the other hand, vigorous, normal-sized plants of the species 
have been observed as far north as Pima, Ariz. 

Although Y. elata is quite variable throughout its range, in several localities 
the variations are characteristic of the majority of plants. These plants may 
possibly represent forms of the species. In Guadalupe Pass, Tex., the plants 
are nanate with stems 0.15 to 1.0 m. tall; leaf blades very rigid and filiferous, 2 
to 8 mm. wide and 15 to 35 cm. long; and flowering stalk 0.7 to 1.5 m. tall. 
The majority of yuccas between Aguila and Salome, Ariz., have very dense 
panicles, with thin, papery capsules. In the vicinity of Pecos, Tex., plants of 
Y. elata commonly have capsules in which the dorsal suture separates at the 
apex, forming a peculiar open pocket. South of Orogrande, N. Mex., and 
elsewhere in sand dune areas, the clumps are quite large, with many heads of 
leaves and comparatively short stems. 

Scattered throughout the New Mexico and Texas ranges of Y. elata, occa- 
sional cases of phyllody (7) occur. They are particularly common in the con- 
centrations east of Van Horn, Tex. 

In 1878 George Engelmann (12) reported Y. elata (Y. angustrfolia var. 
elata) as extending from western Texas to Utah, Arizona, and northern 
Mexico. In 1881 he (13) described Y. elata as having trunks 3 to 5, or even 
11, feet tall, a large-branched panicle, and expanding flowers. At the same 



YUCCAS OF THE SOUTHWEST 61 

time he limited the species range to the deserts of Arizona and probably 
southern New Mexico and made no reference to the Utah plants. The plants 
in the Santa Clara Valley, particularly those near Saint George, were pro- 
visionally referred to as Y. elata by Merriam (26, p. 338) and as Y. radio sa 
(synonym of Y. elata) by Coville (10). 

McKelvey (23) separated the Santa Clara Valley plants as a new species, 
Y. utahensis, and reported its range as, "Plentiful in Washington Co., Utah; 
extending thence into northwestern Arizona and possibly into Lincoln Co.. 
Nevada, near Panaca." 

In Washington, southeastern Iron, and eastern Kane Counties, Utah, the 
yuccas are extremely variable and confusing. The larger part of this area, 
along the Santa Clara and Virgin River basins, is a relatively low (2,500 to 
5,000 feet), warm valley. The soil in the valley varies from a medium-coarse 
gravel to a light sand, which occasionally forms dunes. Although the valley 
is surrounded by mountains frequently covered with snow, snowfall in the 
valley is very light and rarely remains on the ground for any length of time. 
The rest of the area, extending east from the valley through Zion National 
Park into eastern Kane County, is quite mountainous. In eastern Kane County, 
the yuccas extend south from the vicinity 7 of Long Valley (approximately 
6,500 feet elevation) to the Arizona border (around 4,500 feet). Here they 
are mainly on steep mountain slopes, on deep well-drained gravel soil, or 
rarely (10 miles north of Kanab) in sand dune patches. It is the writer's 
opinion that the majority of yuccas in the warmer valley constitute a remote 
stand of Y . elata, which at present is represented by poor or degenerate plants 
and hybrids between the latter species and Y. angustissima. The yuccas in the 
mountainous eastern part of the area are typical Y. angustissima and possible 
Y. angustissima-Y . elata hybrids or derivatives of such hybrids. These hybrids 
extend south from Kanab into Arizona, where only typical Y. angustissima 
occurs. 

A description of Y. elata in the Santa Clara and Virgin River basins is as 
follows: Plant subcaulescent, solitary or small open dump with 3 to 8 heads 
of leaves; stem 0.3 to 2.0 m. tall, trunklike, slender, weak; scape 0.7 to 1.5 m. 
long; flowering stalk 1.5 to 2.5 m. tall; panicle 0.8 to 1.2 m. Jong, ellipsoidal, 
amply branched; sepals 28 to 40 (35.4) mm. long, 13 to 22 (16.6) mm. 
wide; petals 30 to 43 (37.0) mm. long, 18 to 28 (23.5) mm. wide; filaments 
16 to 24 (20.3) mm. long; pistil 23 to 29 mm. long; ovary 5 to 7 (6.1) mm. 
in diameter, stout, pale green or white, oblong-cylindrical, abruptly terminating 
in style, with well-marked carpel sutures and anther depressions; style 7 to 11 
(8.8) mm. long, oblong, very pale green or white; capsule 4 to 6 cm. long, 
2.0 to 3.5 cm. in diameter, oblong-cylindrical, rarely constricted. 

Although as a general rule the average measurements in the plants above 
are somewhat smaller than those given in the preceding description of Y. elata, 
all characters are well within the range of typical Y. elata. As in the case of 
plants in the main Y. elata belt, those plants growing on the coarser gravelly 
slopes are mainly long stemmed, whereas those in the sandy regions are 
shorter stemmed and more clumped. In the Utah range, the former plants 
usually have 1 to 3 weak stems, which are rarely over a meter tall. Along the 
washes, and in the more protected areas, however, plants with stems up to 
1.3 m. tall occasionally occur. A few miles south of Pintura, one exceptionally 
strong, single-stemmed plant was 1.9 m. tall and had 3 branches and 2 old 
inflorescences. The inflorescences were typical of Y. elata and had rather small, 
nonconstricted capsules. In many respects these plants are similar to the weak 



62 AGR. MONOGRAPH 17, U. S. DEFT. OF AGRICULTURE 

ones found along the northern part of the main Y. elata belt in New Mexico. 
The more clumped plants are similar to those found in the sandy regions of 
Orogrande, N. Mex. 

Apparent Hybrids of Yucca elata and Yucca glauca Alliance 

Yucca intermedia var. ramosa McKelvey, Yuccas Southwest. U. S. 2: 120. 
1947. 

Yucca utahensis McKelvey, Yuccas Southwest. U. S. 2: 94. 1947. 
Y. verdiensis McKelvey, Yuccas Southwest. U. S. 2: 98. 1947. 
Y. campestris McKelvey, Yuccas Southwest. U. S. 2: 173. 1947. 
(?) Y. kanabensis McKelvey, Yuccas Southwest. U. S. 2: 122. 1947. 

Along the northern border of its New Mexico and Arizona range and in its 
Utah range, Yucca elata appears to hybridize freely with members of the Y. 
glauca alliance, and plants exhibiting characteristics of both Y. elata and mem- 
bers of the Y. glauca alliance are common. Scattered plants toward the south- 
eastern limits of the range of Y. elata, particularly from Pecos County northeast 
to Howard County, Tex., have characters of both Y. elata and Y. constricta. 
The main transitional zone roughly extends along an imaginary line from Lov- 
ington, N. Mex., to Prescott, Ariz., and varies from 25 to 65 miles in width. 
The majority of transitional forms within this zone may be roughly placed into 
the following groups : ( 1 ) Y. angustissima large, open acaulescent or subcaules- 
cent clump resembling Y . elata, with similar tall flowering stalk and ample, 
branched, ellipsoidal panicle (pi. 51); (2) acaulescent or subcaulescent, soli- 
tary, or small clump, with tall, strong, flowering stalk and long raceme (pi. 
38); and (3) caulescent, solitary, or small clump, with rather short, weak, 
flowering stalk and fairly long raceme (pi. 39). 

Along the Rio Grande basin, the transitional zone is extremely wide and 
clearly shows the transition between Y. elata and Y. glauca (or Y. baileyi). 
Plants with Y . elata characters, or characters apparently influenced by Y. elata, 
are found as far north as Albuquerque; whereas plants showing the influence 
of Y. glauca extend as far south as Las Cruces, N. Mex. The northern Y. 
glauca plants influenced by Y. elata are mainly recognized by the vigorous 
inflorescence that is usually borne above the foliage and somewhat branched, 
and occasionally by a true, ellipsoidal panicle. The plants usually form a small, 
dense clump with only a few heads of leaves, and occasionally they have typical 
Y. elata rhizomes. The Y. elata plants influenced by Y. glauca in the south 
are mainly recognized by the weak inflorescence that is rather low, usually 
somewhat branched, and frequently a true raceme. The plants are usually 
weak and short-stemmed. 

Approximately 22 miles north of Socorro, N. Mex., on a bluff overlooking 
the Rio Grande, is an excellent patch of yuccas. The patch probably extends 
1.5 to 2 miles along the bluff and is about 0.75 mile wide. It is within the 
range of Y. intermedia variety ramosa (p. 58). The plants in this field are ex- 
tremely variable, and neither typical Y . glauca nor Y . elata plants occur. The 
plants are solitary or in small clumps with 2 to 8 heads of leaves. They are 
commonly acaulescent, but occasionally subcaulescent with trunklike stems 30 
to 46 cm. tall. The majority of stemmed plants have large heads of leaves, 
whereas the heads of strictly acaulescent plants are usually rather small. Most of 
the inflorescences are strong and tall, and only 25 percent of them are branched. 



YUCCAS OF THE SOUTHWEST 63 

Among these branched plants slightly branched panicles predominate, but 
panicles similar to those of Y. elata are not uncommon. Measurements in this 
field are as follows: Flowering stalk 70 to 155 (120) cm. long; scape 16 to 74 
(51) cm. long; panicle or raceme 33 to 100 (69) cm. long; sepals 12 to 23 
(18) mm. wide, 28 to 53 (38) mm. long; petals 20 to 32 (27) mm. wide; 
30 to 53 (40) mm. long; filaments 16 to 25 (21) mm. long; pistil 19 to 31 
(26) mm. long; ovary 5 to 11 (7) mm. in diameter; and style 5 to 10 (8) 
mm. 

The average of Y. elata and Y. glauca inflorescence measurements closely 
approaches the measurements of the inflorescences in this field. The measure- 
ments of the flower parts are considerably smaller than those of Y . glauca and 
slightly smaller than those of Y. elata. There is no difference between the size 
of the flowers from panicles resembling Y. elata and from racemes resembling 
Y. glauca. Although the styles are mainly oblong-cylindric, they vary from 
slender to quite stoutish and from very pale green to fairly dark green. The 
capsules average 59 mm. long and 35 mm. in diameter, and approximately 
60 percent of them are slightly constricted. In all respects these plants are 
what would be expected in a field of hybrids and segregating generations of 
crosses between Y. elata and Y. glauca. (Pis. 39, 40, and 41.) 

In central and east-central Arizona, particularly in the vicinities of White- 
river, Show Low, and Springerville, and in the Santa Clara and Virgin River 
basins of Utah, similar apparent hybrids occur. In several characters, however, 
especially in their large open clump and small, deeply constricted capsules, 
these plants resemble Y. angustissima rather than Y. glauca (pi. 51). It is the 
writer's opinion that McKelvey's species Y. verdiensis, Arizona, and Y . utahen- 
sis, Utah, belong to this group of apparent hybrids between Y. elata and Y. 
angustissima. 

McKelvey (25, p. 98) distinguished Y. verdiensis thus: "Plants acaulescent 
to caulescent, forming a single head or small, congested clumps with several 
heads; stem at most short, hidden beneath reflexed leaves . . . Inflorescence 
2.1-2.7 m. in length overall (scape longer than flowering portion) ; inflo- 
rescence proper panicled, raised high above leaves, 1-1.3 m. in length, rarely 
branched throughout, usually with a few long, wide-spreading basal branchiets 
..." McKelvey recognized the similarity between Y. verdiensis and Y. angus- 
tissima and stated (p. 102), " . . . the capsules of both are small, with thin 
wall, smooth epidermis and, at or after dehiscence, the side of the locules 
show a tendency to twist which is not common in other species; the foliage of 
both averages slender. Yucca verdiensis seems to range mainly southeast of Y. 
angustissima and one wonders whether it might perhaps represent a panicled, 
mountain variety of the racemose desert plant. However, its small flowers with 
extremely small and slender pistil and its very long scape, which raises the 
panicled flowering portion high above the leaves, bear no resemblance to those 
of Y. angustissima." 

McKelvey {25, p. 98) also stated, "Yucca utahensis comes closest to Y . 
elata probably, but its often much broader and more concave leaves, non-arb- 
orescent and large clump habit, and, for the most part, smaller flowers with 
usually extremely slender style, separates it from that species." She described 
(p. 94) Y. utahensis thus, "Plants acaulescent or more often with procumbent 
stems 0.6-1.3 m. in length, forming large clumps with 10-15 heads of leaves.' 
She wrote that it prefers a fine, red sandy soil. Of Y. elata in sandy soil near 
Van Horn, Tex., McKelvey reported {p. 93), "... plants were mainly oi 
clump-habit, with numerous short stems and heads of leaves, . . . sending up 



64 AGR. MONOGRAPH 17, U. S. DEPT. OF AGRICULTURE 

many inflorescences, sixteen were counted on one plant." The writer has not 
found the leaves of the plants in the Santa Clara and Virgin River basins to 
be broader or more concave than those found in many Y. elata areas or as 
concaved as the leaves of the plants (Y. kanabensis) in the sand dunes north 
of Kanab, Utah. 

Thinly scattered from Pecos County northeast to Howard County, Tex., and 
northwest into southern New Mexico, and possibly well into northwestern 
New Mexico are plants that appear to be Y. elata-Y. constricta hybrids and 
hybrids segregates. A brief description of these variable plants is as follows: 
Plant an open or rarely rather dense, large acaulescent or subcaulescent clump, 
with stems occasionally a meter in length; head of leaves large, spreading; 
flowering stalk 0.5 to 1.8 m. tall; scape 0.2 to 0.9 m. long, stout; panicle 
usually well above foliage but not uncommonly extending into foliage, ellips- 
oidal, ovoid, or oblate-ovoid; perianth segments 41 to 65 mm. long, 23 to 40 
mm. wide, pistil 26 to 30 mm. long; ovary 5 to 9 mm. in diameter, stout-ovoid, 
pale green; style slender or rather stout, pale to rather dark green; capsule large 
to small, symmetrical or constricted. The open clump of these plants resembles 
that of Y. constricta. Although the stems in subcaulescent plants are con- 
siderably shorter than those of Y. elata, they are two to three times the length 
and much stouter than those of Y . constricta. Although the heads of leaves 
are similar to those of either species, the marginal fibers of the leaves are fine 
and soon erode away as in Y. constricta. The inflorescence is stout, well 
branched, and usually similar in shape to that of Y. elata. Occasionally, how- 
ever, it is broadest toward the base as in Y . constricta. Perhaps one of the 
most confusing characteristics of the plants is the variable height of the panicle, 
which is never as high as in either Y . elata or Y . constricta and which occa- 
sionally extends well into the leaves. The latter character, as well as that of the 
variable color of the style, suggests the possible entrance of Y . glauca into the 
hybrid complex. Such plants are not unlike the possible Y . elata-Y . glauca 
hybrids described in the Rio Grande field (p. 62). Although the flowers of 
the apparent Y. elata-Y. constricta plants could be from either Y. elata or Y. 
constricta, their ovaries more closely resemble Y . elata in size. The capsules 
are frequently as large and symmetrical as those of Y . elata and commonly as 
small and constricted as in many Y . constricta plants. Several miles east of 
Pecos, Tex., the writer found plants resembling Y . constricta with large sym- 
metrical capsules that were indistinguishable from those of Y. elata, whereas 
a few miles east of Pecos, Y. elata plants had capsules similar to those of Y . 
constricta. 

McKelvey (25) separated these apparent Y. elata-Y. constricta hybrids as 
Y. campestris and gave their relationship as close to Y. glauca. 

In the writer's opinion Y. utahensis, Y . verdiensis, and Y. campestris are 
very similar to each other and frequently indistinguishable from many of the 
apparent Y. elata-Y. glauca alliance hybrids. Many of the distinguishing 
characters of the three species are well within the range of the characters of 
Y. elata or members of the Y. glauca alliance. 

Spreading north and southwest from eastern Zion National Park, Utah, are 
plants that appear to be a giant form of Y. angustissima. The inflorescences of 
these plants are mainly racemiform or with a few short abortive branchlets and 
are from 1.8 to 2.5 m. in length. The flowers are large and the ovary somewhat 
larger and style somewhat shorter than those of Y. angustissima. The capsules 
are large, rough, and usually slightly to rather deeply constricted. Although 
these plants commonly form a rather open, small, acaulescent clump, in the 



YUCCAS OF THE SOUTHWEST 65 

sand dunes north of Kanab, Utah, the clumps are large, dense, and frequently 
subcaulescent, with short procumbent stems. 

McKelvey (25, p. 122 J separated the above plants as Y. kanabensis. Speci- 
mens of similar plants have been collected in northern Mohave County, Ariz., 
by R. H. Peebles and Harvey W. Parker, 1940 (pi. 48), and J. Penchney 
Hester 21 reported the plants as occurring from Short Creek to Fredonia, Ariz., 
and to Kave Creek, Utah. Kearney and Peebles (20) referred to the plants 
as a form of V. angusthsima. Although the plants possibly represent a species 
or type, they are very similar to plants occurring in transitional zones between 
Y. elata and Y. glattca alliance, even as far removed as the Rio Grande basin 
(pis. 38 and 39). 

In Texas Y . elata is occasionally associated with, or in close proximity to, Y . 
rostrata, Y. thompsoniana, and Y. reverchoni, whereas in Utah it is close to 
the range of apparent Y . neomexicana-Y . gilbertiana hybrids. No evidence of 
hybrids between Y. elata and these yuccas has been observed, nor has evidence 
been found that Y. elata hybridizes with any of the fleshy-fruited species with 
which it commonly intermingles. 

THE YUCCA MOTH, POLLINATION, 
AND FERTILITY 

The remarkable association of the yucca moths (mainly Tegeticula yuccasella 
(Riley) and Prodoxus quinquepunctellus (Chambers) ) with the yucca plant 
has been known for many years. It is reported that the moth collects pollen from 
several flowers, carries it to another flower, and forces it down the stigmatic 
tube. While forcing the pollen down the stigmatic tube, the moth thrusts its 
ovipositor through the ovary wall and lays eggs. Thus the flower is sure of 
being pollinated and of producing seed, upon which the larvae from the eggs 
feed. Usually the larvae consume only a small part of the seed produced. If 
the association of the moth and yucca proceeds normally, definite stages in 
the life cycles of the two will coincide. Very little data relative to whether 
the yucca depends entirely upon the moth for pollination, whether pollination 
is actually cross-pollination (between two plants), and whether an interference 
in their respective life cycles offset either the moth or plant have been available. 
Since the activities of the moths undoubtedly have an important role in the 
great diversity and distribution of yucca, it appears desirable to record some of 
the writer's observations bearing upon them. 

In yucca concentrations there are usually several yucca moths in practically 
all open flowers, and a single inflorescence generally contains a dozen or more 
moths. Early in the summer of 1942 the writer found that Yucca elata along 
highway U.S. No. 180 from El Paso, Tex., to Carlsbad, N. Mex., was well in 
flower and that the majority of inflorescences as far east as Cornudas, Tex., 
contained numerous moths. East of Cornudas, however, the moths decreased 
in number until, between the vicinities of Salt Flat and Pine Springs, Tex., no 
moths were found. Although from Pine Springs to Carlsbad, N. Mex., Y. 
elata is quite thinly scattered, all inflorescences were amply supplied with 
moths. In the fall of the year the plants were again examined and it was found 
that fruiting occurred only in the areas where the moth was present. 



21 Notes on Hester's specimen No. 678, of leaves and fruit, dated October 1944, in 
U. S. Dept. Agr. Herbarium, Sacaton, Ariz. 



66 AGR. MONOGRAPH 17, U. S. DEPT. OF AGRICULTURE 

In 1942 only a few scattered plants in the dense concentration of Y. glauca 
around Quay, Tucumcari, and Logan, N. Mex., flowered. All of the flowers 
of these plants were heavily infested with yucca moths. In the fall of the year, 
the few flowering stalks were heavily burdened with fruit, but seed damage by 
the moth's larvae was exceptionally high. During the spring of 1943 the con- 
centrations flowered profusely, and, although all inflorescences contained a few 
moths, they appeared to be quite limited in number. Numerous flowers were 
found without moths, and many inflorescences contained only one or two 
moths. In the fall of 1943 it was noted that fruiting was exceptionally light 
in these concentrations. During the years of 1944 and 1945 flowering and 
fruiting in the concentrations appeared to be quite normal. 

In the writer's nursery approximately 15 plants have flowered, and each was 
in flower at the same time that 3 or more other plants were flowering. There 
are no yucca moths in the nursery, and, except in the case of Y. whtpplei and 
of hand-pollinated flowers, no fruits set. Table 2 gives the results of hand- 
pollinations of several yucca species and forms in the nursery. 

Table 2 represents 977 pollinations, of which 21.51 percent set fruit con- 
taining numerous apparently viable seed. In both selfed and cross-pollinated 
plants, the fertility varied from to 93.34 percent. A total of 225 self- and 
752 cross-pollinations were made, which, respectively, set 24.89 and 21.81 
percent fruit. Of the pollinations between apparent hybrids 53.33 percent set 
fruit, while only 6.57 percent of the pollinations between typical species set 
fruit. Seventeen percent of the pollinations between apparent hybrids and 
species set fruit. Although these percentages indicate that the plants are about 
equally self- and cross-fertile, it is evident that the apparent hybrids are con- 
siderably more fertile than the species. 

Table 2 reveals that Y. arizonica, Y. neomexicana, Y. data, Y. reverchoni, 
Y. rupicola, Y. thompsoniana, and the apparent hybrids Y. thompsoniatia P 
3 and Y. glauca 11-2 are self -sterile or nearly so. Although Y. neomexicana, 
Y. reverchoni, Y. rupicola, Y. thompsoniana P 1, and Y. el at a are nearly 
female-sterile, they are highly male-fertile. Y. schidigera and the apparent 
hybrids Y. glauca 11-1 and Y. constricta are rather highly self-fertile, whereas 
the last two plants and apparent Y. thompsoniana P 3 hybrid are just as highly 
male- and female-fertile. 

There appears to be little or no correlation between the relationships of the 
species and the results of cross-pollinations. In the majority of crosses between 
the baccate- and capsular-fruited yuccas, the baccate-fruited species was mainly 
used as the staminate parent. The only crosses employing the baccate- fruited 
species as the pistillate parent are Y. arizonica-Y . glauca apparent hybrid and 
Y . arizonica-Y . neomexicana; 50 and 70 percent, respectively, of these crosses 
set fruit. Within the capsular-fruited species Y. neomexicana crosses readily 
with the distantly related apparent hybrids of Y. glauca and Y. constricta, and 
Y. thompsoniana likewise easily crosses with remotely related Y. elata-Y. 
glauca apparent hybrid, and Y. constricta apparent hybrid. Trelease (31) 
cited several garden hybrids within the capsular-fruited species Y. filamentosa 
and lists several artificial hybrids between the latter species and the baccate- 
fruited species Y. gloriosa. 

Although several of the yuccas listed in table 2 commonly or always have 
constricted fruit, none of the hand-pollinations resulted in constricted fruit. 
Unquestionably the yucca moth is responsible for the occurrence of constricted 
fruit in native plants. It is very probable that the constrictions are caused by 
the failure of the injured ovules to develop. 



YUCCAS OF THE SOUTHWEST 67 

In the spring of 1950 two plants of Y . whipplei obtained from the badlands, 
northwest of Eden Hot Springs, Riverside County, Calif., flowered in the 
nursery. These plants were adjacent to each other and flowered simultaneously. 
Throughout the flowering period no yucca moths were found in the nursery. 
One of the plants fruited profusely, and every capsule was completely filled 
with seed. No capsule or seed showed any signs of the yucca moth or larvae 
having been present. Pollination had undoubtedly been caused by insects 
other than the yucca moth, which were abundant in the flowers. Although it is 
not known how general such pollination occurs in Y. whipplei, it is very likely 
quite common, as Y. whipplei is the only yucca with a capitate stigma. The 
second plant failed to set fruit and died shortly after flowering. On several 
occasions the writer has observed similar fruiting and nonfruiting plants of 
Y. whipplei in the badlands section, as well as in the foothills on the San 
Bernardino and the San Gabriel Mountains. The cause of the apparent sterility 
is unknown. 

Of the flowers of the yuccas in table 2, only the hand-pollinated ones set 
fruit, which indicates that many if not all of our southwestern yuccas except Y. 
whipplei depend upon the yucca moth for pollination and consequently the 
development of fruit and seed. When the yuccas over a wide range flower 
sparsely, it is apparent that the moth's lack of breeding places and the limited 
food supply for the larvae greatly reduces the moth population. Possibly the 
lack of moths between Salt Flats and Pine Springs, Tex., in 1942 indicates that 
the plants failed to flower in 1941, and the moth population in the region was 
obliterated. Following a reduction in moth population, it is apparent that 
recovery to a balanced condition of flowering, fruiting, and moth infestation 
would depend upon the degree of reduction and the size of the area involved. 
However, mothless or nearly mothless areas covered a relatively small part of 
the range of a yucca species, and it is very likely normal flowering, fruiting, and 
moth infestation would be recovered in several years. 

Outside of Y. brevifolia and Y. whipplei, our southwestern yuccas are not 
reproducing to any extent by seeds. There can be little question, therefore, 
that the yucca moth is more dependent on the yucca for its existence than the 
yucca is on the moth. During their long life through vegetative reproduction, 
the majority of yuccas would continue to exist for many years without the 
moth. On the other hand, it appears that the yucca moth would be completely 
wiped out if the yuccas failed to flower for a single year. 

Regardless of the fact that yuccas are about equally self- and cross-fertile 
and that the moth flies from flower to flower, it is doubtful if cross-pollinations 
are as prevalent as reported. It is very likely that the number of self-pollinated 
flowers far exceeds the number of cross-pollinated ones, and except in areas 
where species are admixed and flower at the same time interspecific crossing is 
but remotely possible. 

YUCCA SEEDS AND SEEDLINGS 

The seeds of southwestern yuccas may roughly be separated into four mor- 
phological types, which, in the majority of species, are characteristic of the 
sections of the genus. In the indehiscent-fruited species the seeds of section 
Sarcocarpa are flat, thick (2 to 3 mm.), rough-surfaced, wingless, and oval or 
circular in outline, whereas those of section Clistocarpa are flat, thin (rarely up 
to 1.5 mm. thick), smooth-surfaced, wingless, and irregularly oval in shape. 



68 



AGR. MONOGRAPH 17, U. S. DEPT. OF AGRICULTURE 












6" o 4 



CO 




w 


v 


-•-} 

8 


| 


co 




! 


^ 

Q 


<N 


GQ 


h3 


b 


- 




< 




H 





V1VJ9 \l 



npoylnj, • £ 



mo^oidadi ^ 



z d i 'q^q 



t z i. -q^q 



£ d i q A q 

VUDlU0SdlU0l[1 ■ x 



I d 

miviuosdwom '[ 



VUOJIXdUlOdU 



S-6T 'ON 



T-6T 'o^r 



S-II/ox 



i-ti -m 



vjdbippps ' x 



r» to 

CO rf< 

CO tH 

CO *Q 



oo 1 O cO 

0^ (M 



g ^VOIUOZIMD '£ 



I (fvoiuozuv ' £ 



£° 2 



8J 






S b ! b ! b ! b 

I I* JS is M 

1 i|-|oioio 



I S 



k 



>, >h ^ ^ 



>H >* 



YUCCAS OF THE SOUTHWEST 



69 



CO 
d 
co 






1 CO 

1 <£> 










1 TfH 




i Q «-0 

; d ^' 


! o 


86.67 

60.00 



i co 


93.34 
93.34 








i 










54.45 
40.00 




o 


CO 

d 


























o 






73.34 
6.67 










o 
o 
d 








r»5 O 
































If 


He- 

1-3 


1 

1 


•I 


1 





0. 
"^ si ai to m no 

fl a a a a 
o.o o o o 

co '-3 +3 VE> *^3 V3 
C 03 c3 c3 ed oj 

.2 3 3 3^1 
■43 adg^a 

3 . o o o'p'o 

1-1 a G, Oh Ch Q- Qi 
2 O -73 T3 T3 T3 T3 

£fi.h 4) CU <D QJ CU 

Id**- 1 co eg co co 35 

-rt 03 C3 03 cj cd 
CO.g^ CO ^. ifl w 



t3 'S "^ CO ^ "^ - CU 

CU 3 Ph<-! O) K J? 

73 O ££ §tt C3 ^ g 



03 a 



P43^D 



•° ff-S 8 2 9 * w 






*CD 

eJfl 

co 



£3 J 



S co „_ co O 

* cd CD CQ-3 

^ fe -,£3 CU Cd CD CO 

4^s-g| .§g 

3-<cu^'^cucucu 

3 8| grigs 8. 

- . .SPSS =d a = 



70 



AGR. MONOGRAPH 17, U. S. DEPT. OF AGRICULTURE 



In the dehiscent-fruited species the seeds of section Hesperoyucca are fiat, ex- 
tremely thin and papery, smooth-surfaced, wingless, and irregularly ovate or 
oval in outline. With the exception of members of the Yucca rupicola alliance 
the seeds of section Chaenocarpa are similar to those of Hesperoyucca, but 
they possess a marginal wing that is exceptionally wide or conspicuous at the 
chalazal end of the seed. The seeds of Y. rupicola alliance are similar to 
those of other members of the Chaenocarpa, but their marginal wing is ex- 
tremely narrow or wanting. 

In general yucca seeds consist of hard, tough seed coats, an abundant horny 
endosperm, and a comparatively small embryo. The embryos of Y. brevifolia 
and the dehiscent-fruited species are straight, terete, and elongated, whereas 
those of the indehiscent-fruited species are somewhat shorter and more conical 
or subconical in shape. The cotyledonary leaf of the embryo is very large in pro- 
portion to the axis. The rudiment of the primary root is at the posterior end ot 
the axis, and in the majority of species it is difficult to distinguish it from the 
cotyledon end. The plumule is undifferentiated or is distinguishable only by a 
dense meristematic region within the cotyledon at the anterior end of the axis. 



Germination 



The more or less waxy, tough, outer seed coat, the abundant endosperm, 
and the high vitality of yucca seeds result in their being well adapted to semi- 
arid deserts. In germinating tests by the writer, the seeds were soaked in water 
for 24 hours, then placed between folds of moist cotton, and kept at tempera- 
tures between 28° and 32° C. Under these conditions germination was fairly 
uniform and was as follows: 



Species 

Yucca brevifolia 

Y. whipplei 

Y. elata 

Y. glauca 



Germination Germination 

(Days) Species (Days) 

3 Yucca torreyi 6 

3 Y. schidigera 6 

4 Y. faxoniana 7 

4 Y. rupicola 9 

In the majority of tests, the germination periods were greatly reduced by 
carefully scarring or removing the seed coats at the hilum end of the seeds. 

Out of approximately 75 germination tests, only 3 samples of seeds failed 
to show any germination. Two of these samples were from apparent hybrids 
between Yucca glauca and Y. elata, collected along the Rio Grande north of 
Socorro, N. Mex., and the other from a possible Y. angustissima-Y . elata hybrid, 
collected near Whiteriver, Ariz. The viability of the rest of the seed samples 
varied from 45 to 98 percent, with the majority ranging from 80 to 90 percent. 
Seeds of high viability were collected from fairly young, green fruits as well 
as from 2 -year-old fruits, which were considerably weather-beaten. One lot of 
Y. elata seeds collected in 1942 showed approximately 95 percent germination 
in 1948. 

In the semi-irrigated nursery at Riverside, Calif., the germination period of 
southwestern yucca seeds is spread over several years. When the seeds are 
planted in the early spring the first germination usually occurs within 1 or 2 
weeks. This is followed by continuous germination of other seeds, until at 
the end of 2 or 3 years the plot is composed of seedlings varying from several 
months to several years old. 



YUCCAS OF THE SOUTHWEST 71 

The first stage of yucca seed germination is the pushing out of the radicle 
at the base, or hilum end, of the seed. Apparently this appearance of the 
radicle is caused by simultaneous growth of the cotyledon and the radicle. At 
the end of 2 or 3 days the cotyledon has elongated to such an extent that the 
axis is well removed from the remains of the old seed. During this period, 
the primary root becomes well denned and the differentiation of the plumule 
within the cotyledon becomes more apparent. At the end of 4 days, the 
cotyledon has reached its maximum length; the plumule, still enveloped in 
the now swollen cotyledon, is well developed; and the primary root is con- 
siderably elongated and has root hairs. When the developing seedling is 
approximately 7 days old, the plumule breaks through the cotyledon (see 
pi. 62). 

As a general rule, seed germination and seedling development of the cap- 
sular-fruited species is more rapid than that of the fleshy-fruited species. The 
length of cotyledon growth of the capsular- fruited species (see pi. 62) is much 
greater than that of the fleshy- fruited species (pi. 52). The longest cotyledon 
extensions measured for these groups were: Y. elata — capsular, 6.5 cm. 
long; and Y. torreyi — fleshy-fruited, 3.4 cm. long. In soil the cotyledon ex- 
tension frequently forms an arch above the soil surface, thus removing the 
young seedling from the old seed remnants. As the seedling develops, it 
gradually becomes adjacent to the old seed and the cotyledon forms a pro- 
tective sheath around the base of the plumule (pis. 52 and 62). In several 
cases the cotyledonary arch was broken soon after its formation, thus separating 
the young, undeveloped axis from the remains of the old seed. In all of 
these cases, the axis continued to develop and exhibited no adverse effects. 

An early work (1) described the arch cotyledon as containing the complete 
embryo and concluded ". . . the embryo is thrown out previous to germina- 
tion . . ." The formation of the cotyledonary arches in yucca and other mono- 
cotyledonous plants has caused some speculation as to growth reactions and 
tropisms involved. According to Trelease (31) the subterete elongated embryo 
obliquely placed across the seed and the arched cotyledon are characteristics 
of Yucceae. 

Seedlings 

Although southwestern yuccas produce an abundance of viable seed, repro- 
duction by seed is very limited. During the writer's 4 years of field work he 
has recorded the following numbers of seedlings: 

Species of more arid environs Seedlings : Species of less arid environs Seedlings 

Yucca elata 29 Yucca glauca 72 

Y. baccata (southern N. Mex.).. 9 Y. neomexic.ina 52 

Y. torreyi 8 Y. rupicola 17 

Y. arizonica 7 Y. baccata (northern N. Mex.). .16 



Y. constricta 5 

Y. reverchoni 3 

Y. schidigera (Arizona) 1 

Y. faxoniana 

Y. carnerosana 

Y. schottii 

Y. rostrata 

Y. thompsoniana 



Y. baileyi 13 

Y. angustissima 6 

Y. schidigera (southern Calif.).. 6 

Y. brei-ifolia 0) 

Y. whipplei ( x ) 



1 Unlimited. 



72 AGR. MONOGRAPH 17, U. S. DEPT. OF AGRICULTURE 

In a relatively dense stand of Yucca elata in southern New Mexico, Campbell 
and Keller (8) reported that in 17 annual chartings of 75 meter quadrats, 
only 10 seedlings were observed; and that in 1 year only 6 seedlings were 
recorded in 170 quadrats. 

The paucity of yucca seedlings can undoubtedly be attributed to meager and 
irregular rainfall, to extremely slow growth and establishment of seedlings, 
and to rodent damage. The preceding list of species and seedlings are roughly 
placed into more or less arid groups. In the less arid group there are con- 
siderably more seedlings than in the more arid group. On the United States 
Jornada Experimental Range, near Las Cruces, N. Mex., where the mean 
annual rainfall is approximately 8.24 inches, the writer has recorded 7 Y. elata 
seedlings; whereas at the United States Department of Agriculture Field 
Station near Tucumcari, N. Mex., where the annual precipitation is approx- 
imately 16.42 inches, he found 46 Y. glauca seedlings. The viability and 
germination period of the seeds of Y. elata and Y. glauca collected at these 
localities are approximately equal. That the amount of rainfall is partly 
responsible for the lack of yucca seedlings is indicated by the distribution of 
Y. brevi folia seedlings in the Mojave Desert of California. In the lower 
valleys and mesas of this desert, the annual precipitation is meager and there 
are very few seedlings of Y. brevifolia, but in the higher mountain areas the 
rainfall is much greater and Y . brevifolia seedlings are plentiful. Y. brevifolia 
seedlings are particularly common in the vicinities of Yucca Valley and 
Twentynine Palms on the northern slopes of the Little San Bernardino 
Mountains. 

It is very likely that yucca seeds often remain in the soil for several years 
and mainly germinate during the favorable seasons, which may be many years 
apart. On several occasions the writer has observed many seedlings of ap- 
proximately the same age, but older or younger seedlings were lacking. In 
June 1945 approximately 20 seedlings of Y. baileyi var. navajoa were found 
in a relatively small area north of Tohatchi, N. Mex. Although these seed^ 
lings had only succulent, juvenile leaves, their subterranean growth indicated 
they were approximately iy 2 to 2 years old. Five miles south of the area in 
another fairly dense concentration of variety navajoa no seedlings were 
found. Similar patches of seedlings of Y. elata have been observed near Silver 
City, N. Mex., and of Y. gilbert iana in the House Range of Utah. In these 
places the rainfall is sporadic. 

Yucca seedlings are extremely slow in growth, and they retain their grass- 
like, often succulent, juvenile leaves for fully a year. Since the forage is 
scarce in the majority of the desert areas, such rodents as wood rats and rabbits 
habitually cut off the seedling leaves. Consequently, seedlings are often found 
that have characteristic shoots of 6- to 10-month-old plants, but their sub- 
terranean development indicates they are actually several years old. 

In one of their meter quadrats, Campbell and Keller (8) charted a single 
Y. elata seedling for 6 years, when it disappeared. After 3 years had elapsed, 
the seedling reappeared and continued to grow. The authors attributed the 
plant's disappearance to rodent damage. In many of the drier parts of the 
southwestern deserts, grazing by sheep and goats is particularly damaging to 
yucca seedlings. In areas of higher rainfall and, consequently better forage 
conditions, yucca seedlings develop normally and rarely exhibit evidence of 
rodent damage. 



YUCCAS OF THE SOUTHWEST 73 

Aerial Characters 

The juvenile leaves of yucca seedlings are commonly succulent, flaccid, 
and usually a glaucous blue green. The seedlings of capsular-fruited species 
and of Yucca brevijolia have numerous, usually narrow, and markedly grass- 
like leaves (pis. 53 and 62), whereas the seedlings of the fleshy-fruited 
species have comparatively few, generally broad, rather reedlike leaves (pi. 52). 

When the seedlings become 4 to 6 months old the juvenile leaves are 
gradually replaced by small maturelike leaves, which in turn are gradually 
replaced by typically matured leaves (pis. 52 and 62). This transformation 
requires from 18 months to 3 years. 

Following the formation of matured leaves, some of the species are distin- 
guishable in their seedling stage, but others are difficult or impossible to 
separate. The seedlings that are rather easily separated and the characters 
distinguishing them are: 

Yucca, jaxoniana, with leaves comparatively short, broad, deeply concave, dark green, 

and short, coarse, curly marginal fibers. 
Y. schottii, with leaves thin, slightly concave, light glaucous green, and without 

marginal fibers (pi. 52). 
Y. arizonica, with leaves narrow and a few fine, straight marginal fibers (pi. 54, B) . 
Y. baccata-Y. arizonica (Y. thornberi McKelvey) (Superior, Ariz.), similar to Y. 

torreyi and Y. baccata but with leaves more glaucous blue green and somewhat 

longer, less curled marginal fibers. 
Y. torreyi (from west-central Texas), with leaves long, broad, light green and 

rather long, fine, slightly curled marginal fibers. 
Y. brevijolia, with head of numerous ascending leaves; leaves short, narrow, stiff, 

blue green, and serrate margins. 
Y. brevijolia var. jaegeriana, with head of numerous spreading leaves; leaves very 

short, narrow, stiff, yellowish green, and serrate margins. 
Y. whipplei, with leaves long, narrow, rather flaccid, drooping, dark blue green, 

with serrate margins. 
Y. whipplei var. parshii ?, with leaves long, broad, stiff, spreading, wavy, glaucous 

blue green, with serrate margins. 
Y. whipplei var. caespitosa, with numerous auxiliary leaf heads, leaves similar to 

those of species. 
Y. thompsoniana, with head of numerous leaves; leaves rather long and narrow, 

stiff, spreading, yellowish green, with horny, yellow, serrate margins. 
Y. reverchoni, with head of few leaves; leaves short, narrow, stiff, spreading, 

yellowish green, with horny, yellow serrate margins. 
Y. rupicola, with head of very few leaves; leaves long, broad, rather flaccid, twisted, 

wavy, with horny, yellow margins and brown serrations. 
Y. neomexicana, with leaves narrow, straight, spreading, with few fine .straight 

fibers. 
Y . gilbertiana, with leaves broad, wavy, somewhat falcate, with numerous coarse, 

curly fibers. 
Y. baileyi var. navajoa (Tohatchi, N. Mex.), similar to Y. glauca, Y. baileyi, Y. 

constricta, and Y. elata, but with leaves shorter and more filiferous. 
Y. elata (Vail Junction, Ariz., State College and White Sands National Monument, 

N. Mex.), similar to Y. glauca, Y. baileyi, and Y. constricta, but with consider- 
ably more leaves than equally as old seedlings of the other species. 

All of the above seedlings comprise fairly uniform populations. Single 
plant progenies composed of two or more types of seedlings are: 

Y. torreyi (State College, N. Mex., and El Paso, Tex.), with leaves either (1) large, 
straight, spreading; or (2) small, skewed-falcate. 

Y. torreyi (Sonora and Sterling City, Tex.), with leaves either (1) large, straight, 
spreading, rather light green; or (2) a typical blue form (p. 23); or (3) an in- 
termediate type. 



74 AGR. MONOGRAPH 17, U. S. DEPT. OF AGRICULTURE 

Y. baccata (Mountain Park, N. Mex.) either (1) vigorous, with leaves straight, 
spreading, rather light green, and quite coarse, curly marginal fibers; or (2) very 
small, with leaves skewed-falcate, dark green, and very coarse, curly marginal 
fibers; or (3) an intermediate type. 

Y. whip pi ei hybrid? (Cajon Pass, San Bernardino County, Calif.), either (1) 
solitary, with leaves long, narrow, stiff, straight, spreading, and very glaucous 
blue green; or (2) caespitose, with leaves long, wide, falcate, somewhat flaccid, 
spreading, and very light, glaucous blue green; or (3) an intermediate type (pi. 
55). When these seedlings were 15 months old they segregated as follows: 20 
seedlings solitary with broad, very glaucous, light blue-green leaves, and 2 seed- 
lings, caespitose, with 8 and 14 secondary leaf heads, derived from buds in axes 
of seedling leaves; leaves rather narrow, stiff, straight, slightly glaucous, dark 
blue green. 

Single plant progenies with types distinguishable only or mainly by repro- 
ductive characters are: 

Y. tbompsoniana-Y. rererchoni? (Sheffield, Tex.), seedlings either (1) Y. thorn p- 
soniana type with numerous long leaves, long flowering stalk, and large, broad 
panicle; or (2) Y. reverchoni type, with few short leaves, short flowering stalk, 
and small, narrow panicle; or (3) a quite variable intermediate type. 

Y. glauca hybrid? No. 11 (Tucumcari, N. Mex.), seedlings with either (1) panicle 
narrow with dorsal side of branchlets forming a narrow angle with inflorescence 
stalk, pistil roughly glandular, style tumid with basal lobes protruding over ovary, 
capsules very rough and glandular, anther depressions large and deep; or (2) 
panicle broad with dorsal side of branchlets forming a wide angle with inflo- 
rescence stalk, pistil smooth, style slender, not lobed, capsules smooth, no glands 
or anther depressions. 

Y. glauca hybrid? (Y. intermedia McKelvey) (Grant, N. Mex.), inflorescences 
either (1) strictly racemose (pi. 42); or (2) slightly branched; or (3) one 
plant, very likely owing to teratological variation, broadly paniculate. The in- 
florescence bud of this plant developed very slowly and produced very short, broad 
leaves that terminated in a large, pungent, incurved hook. When the foliaceous 
inflorescence was 11 months old (82 cm. long) the bud became more active and 
rapidly developed a normal, broad, ellipsoidal panicle that produced normal 
flowers, capsules, and seeds (pi. 43). 

Y. constricta-Y . elata? (Y. campestris McKelvey) (Big Springs, Tex.), seedlings with 
either (1) scape short, stout, panicle extending into foliage, narrowly ellipsoidal, 
branchlets very ascending,, flower green, style stout, swollen at base, and pro- 
truding over ovary; or (2) scape long, weak, panicle well above foliage, broadly 
ellipsoidal, branchlets barely ascending, flower white or greenish white, tinged 
with purple, style slender, not protruding over ovary. 

Progenies of field selected for definite characters are: 

Y. schottii (Patagonia, Ariz.), (1) all seedlings of selected longleaf plant had long 
leaves (average 71.7 cm.); (2) all seedlings of shortleaf plant had short leaves 
(average 50.7 cm.). 

Y. glauca hybrid? (Tucumcari, N. Mex.), field plant selected for long leaf and 
large, broad panicle; all seedlings had long leaves, but segregated into two types 
of panicles (see Y . glauca hybrid No. 11 above). 

Y. elata (Vail Junction, Ariz.), field plant selected for long, drooping or skewed- 
falcate leaves, one seedling had leaves similar to parent, and five seedlings had 
comparatively narrow, short leaves. 

Y. elata (White Sands National Monument, N. Mex.), all seedlings from selected 
longleaf plant had long leaves (average 101.5 cm.). 

Y. elata (State College, N. Mex.), all seedlings from selected shortleaf plant had 
short leaves (average 49.0 cm.). 

Only one lot of seedlings of Y. schidigera (Twentynine Palms, Calif.) has 
been grown. These seedlings had rather short, skewed-falcate, dark glaucous 
blue-green leaves. They were very similar to the skewed-falcate-leaved plants 
of Y. torreyi from State College, N. Mex., and El Paso, Tex., and of Y. 
baccata from Mountain Park, N. Mex. 



YUCCAS OF THE SOUTH W F.ST 75 

It is very likely that the skewed-falcate-leaved plants of Y. baccata constitute 
an undescribed type, differing from the species and its variety vespert'tna 
in its small dump of few leaf heads, and its very short, skewed -falcate dark- 
green stiff leaves. The writer has noted such plants thinly scattered on the 
rolling, dry grassland hills in west-central New Mexico. 

In the seedlings listed above, the major genetical variations occur either 

(1) as a progeny of uniform seedlings, all differing from the seed parent; or 

(2) as a progeny of several types of individuals. In the first group are the 
uniform straight, spreading-leaved progeny and the uniform small, skewed - 
falcate-leaved progeny from typical Y. torreyi occurring at State College, N. 
Mex., and El Paso, Tex. Possibly belonging to the same group are the 
typical, blue-green, and intermediate progenies from typical Y. torreyi occur- 
ring at Sonora and Sterling City, Tex. It is very likely that these variations 
are a result of hybridization, since the straight-leaved seedlings are indis- 
tinguishable from Y. torreyi seedlings and the skewed-falcate-leaved seedlings 
resemble similar seedlings of Y. baccata. In the second group, or progeny of 
several distinct types of individuals, are the seedlings of Y. wbipplei from 
Cajon Pass; Y. baccata from Mountain Park; Y. thompsoniana from Sheffield; 
Y. glauca from Tucumcari and Grant; and Y. con strict a from Big Springs. 
It has been shown that the Y. ivhipplei seed parent was probably an intra- 
specific hybrid between varieties parishii and caespitosa, and that the seedlings 
segregated into the two varieties, as well as the hybrid form. The seedlings of 
Y. baccata probably also represent an intraspecific hybrid segregation, since they 
are of three types, a typical Y. baccata, a. small skewed-falcate-leaved form, 
and an intermediate form. The Y. thompsoniana seedlings were from a plant 
noted to be a possible hybrid with Y. reverchoni. The progeny segregated into 
forms resembling Y. thompsoniana, Y. reverchoni, and a form midway between 
these species. Y. glauca apparent hybrid from Grant segregates into a racemi- 
form type and a paniculate type. Y. glauca from Tucumcari and Y. constricta 
apparent hybrids segregate into distinct inflorescence and flower types. 

Although many of the variations that occur in native yucca fields are 
undoubtedly caused by environmental conditions, such progenies as the short- 
leaf and longleaf selections of Y. schottii show that variations are frequently 
caused by minor genetical differences. These differences undoubtedly account 
for the variations in leaf size and habit of the various progenies of Y. torreyi, 
Y. elata, and Y. glauca. 

Subterranean Characters 

The primary roots of young yucca seedlings are characteristic of mono- 
cotyledonous seedlings. When they are from 10 to 18 cm. long, they are 
succeeded by lateral roots (pis. 52, 53, and 62) that develop from the axis. 
These roots soon become very swollen and more or less fusiform in shape. 
They are covered with a smooth, yellowish- or reddish-brown epidermis and 
have a typical protostele root structure. The stele and endodermis are very 
small in comparison with the thick layers of cortical parenchyma cells. The 
latter tissue is very succulent, and upon slight pressure it readily collapses and 
separates from the relatively hard endodermis and stele. 

At the preceding stage of development, or when the seedlings are approx- 
imately 3 to 5 months old, only one species is distinguishable by its subter- 
ranean growth. This species, Yucca whipplei, forms a definite, swollen, bulb- 



76 AGR. MONOGRAPH 17, U. S. DEPT. OF AGRICULTURE 

like, undergrowth stem; whereas in all other species this bulbous appearance 
is very slight or wanting (pi. 53). 

All species continue to form secondary roots until the seedlings are approx- 
imately 6 to 8 months old, when, with one or two exceptions, the dehiscent- 
fruited forms follow a rhizomatous development (pis. 62 to 72), whereas the 
indehiscent-fruited species form a typical fibrous root system (pis. 52 to 61). 

Four-year-old seedlings of all indehiscent-fruited species have a well-devel- 
oped fibrous root system. These roots spring from the rounded, somewhat 
swollen base of the seedlings and spread obliquely into the soil. At approx- 
imately this stage of development, the seedlings of Y. arizonica and Y. baccata 
(pi. 54) form from 1 to 4 fairly large nodules just below the soil surface in 
the region of the young seedling axis. These nodules extend from 6 to 10 
cm. from the seedling, and are covered with a thin, rough bark. On their 
undersurface, succulent fusiform roots develop, and at the tip of each, a 
large leaf-head bud is produced. In only one other indehiscent-fruited species 
has such leaf-head formation been observed in the seedling stage. In this 
case, 7 out of 10 Y. torreyi seedlings, from seed collected in the vicinity of 
State College, N. Mex., produced a single head of leaves from a rhizomelike 
nodule. 

Five- to eight-month-old seedlings of the capsular-fruited yuccas exhibit a 
single, relatively short primary root and from one to four long, succulent, 
fusiform, secondary roots. The primary root is seldom branched, but the 
secondary roots usually have from one to three branches toward their ends. 
Both types of roots are covered with a smooth, thin, reddish- or yellowish- 
brown epidermis and have a protostele structure. Except for Y. wbipplei, 
when the seedlings are between 6 and 9 months old, a large swelling occurs 
on the axis and soon develops into a scale-covered rhizome bud (pi. 62). 
Although only one bud usually forms, occasionally two or three are con- 
currently developed. The rhizome develops very slowly and is covered with 
a thick, smooth epidermis. It has a dictyostele structure, lacks leaf scales, 
and terminates in a large blunt, scale-covered bud. From the rhizomes, fine, 
branched roots and large fusiform adventitious roots develop. The latter 
rhizome roots are indistinguishable from the secondary axis roots, which they 
soon replace. They have a protostele structure. In all species, except Y. elata. 
the rhizomes soon branch (pi. 63) and mainly develop obliquely into the 
soil. In Y. elata seedlings, the rhizome remains unbranched and grows ver- 
tically downward (pi. 64). 

Three- to four-year-old seedlings of Y. wbipplei and its varieties parishii 
(?) and caespitosa do not form rhizomes. Equally as old seedlings of appar- 
ent hybrids between the latter varieties appear to have short rhizomes, which 
ascend obliquely from the seedling axis region and terminate in a single 
leaf head. The writer has not grown seedlings of Y . wbipplei var. percursa, 
and he has examined only very young seedlings of Y. rostrata, Y. tbompson- 
iana, and Y. angustissima. 

In the preceding taxonomic section, it was shown that many of the tran- 
sitional forms of Y . glauca varied considerably in the number of leaf heads 
per square meter, and, in general, the plants of western New Mexico were 
less dense than those of eastern New Mexico. Such density undoubtedly is 
partly caused by either a combination of the frequency of rhizome branching, 
the length of rhizome branches, or the frequency of rhizome leaf-head forma- 
tion. That a possible correlation exists between the amount of rhizome 



YUCCAS OF THE SOUTHWEST 77 

branching and plant density is indicated by the following differences in 
rhizome behaviors: 

Yucca glauca hybrid? (Tucumcari, N. Mex.). — Seed parent density, 11.2 
heads of leaves per square meter; 71.43 percent of the rhizomes of 15-month- 
old seedlings had from 3 to 5 branches. 

Yucca glauca hybrid? (Grant, N. Mex.). — Seed parent density, 5.5 heads 
of leaves per square meter; 50.02 percent of the rhizomes of 15-month-old 
seedlings had from 2 to 4 branches. The rhizomes in these seedlings were 
approximately the same length. 

ROOT AND SHOOT CHARACTERS OF MATURE 

YUCCAS 

Indehiscent-Fruited Species 

In matured field plants the base of the trunklike stem is either flat (in Yucca 
carnerosana, Y. faxoniana, and Y. brevijolia, and rarely in Y. scbottii, Y. tor- 
reyi, and Y. scbidigera) (pis. 56 and 57), or rounded (in Y. arizonica and 
Y. baccata, and commonly in Y. scbottii, Y. torreyi, and Y. scbidigera) (pi. 
58). Except in Y. brevijolia, the margins of the flat bases are wavy in outline 
and rounded under, giving a folded appearance. These rounded margins ex- 
tend several centimeters into the soil and they appear to be composed of 
highly active tissue. Just inside of the rounded margin, a band of fibrous roots 
encircles the stem base (pi. 56). The rounded or hemispherical stem bases 
extend 20 to 40 or more centimeters into the soil. Although the roots of these 
plants are mainly limited to the outer regions of the rounded base, they do not 
occur in a definite band. The roots of all plants are long and tough. In typical 
flat-based species, the roots are less branched and slightly larger in diameter 
than those of the species with rounded bases. 

In Y. carnerosana, Y . jaxoniana, Y . scbottii, Y . torreyi, and Y . scbidigera 
basal sprouts develop from nodules formed in the active tissues at the base 
of the stem (pi. 57). These nodules are formed just below the soil surface 
and are usually hemispherical in shape. They are covered with a rather thick, 
rough bark. A few fibrous roots develop in the under surface of the nodules, 
and on their upper surface, usually toward the center, a sprout bud is formed. 

The nodules are always close to the soil surface, and, except in Y. torreyi and 
Y. scbidigera, no subterranean stem extensions other than the nodules them- 
selves have been observed. The nodules of Y. torreyi and Y. scbidigera occa- 
sionally extend 15 to 20 cm. from the parent stem (pi. 58). In no case, 
however, have these extensions been found to have terminal buds, and sprout 
formation from them appears to be entirely adventitious. 

Yucca carnerosana, Y. faxoniana, and Y, scbottii commonly have 1 or 2 
and rarely up to 5 basal sprouts. All sprouts are produced at the base of the 
stems and are contiguous to the mother stems. Plants having 2 or 3 equally 
tall stems are believed to have developed either from sprouts of an old stem 
that no longer exists or from multiple-stemmed seedlings. Y. torreyi and Y. 
scbidigera commonly have 1 to 5, and rarely up to 10 or more sprouts, which 
are adjacent, or close to the mother stems. 

Slightly north of Stoddard Well, in May 1943, the writer found a large 
burned grassland area, containing numerous plants of Y. scbidigera. It 



78 AGR. MONOGRAPH 17, U. S. DEPT. OF AGRICULTURE 

seemed probable that the fire had occurred the previous summer. Upon ex- 
amination of the Y. schidigera plants, only a comparatively few were found 
to show any signs of life and it was concluded that practically all plants had 
been killed by the fire. In May 1944 the plants were again examined and it 
was found that although all of the old shoots had been killed to, or nearly to, 
the ground, practically all plants had formed one to four sprouts at the 
bases of the old burned stems. Although a few plants showed no signs of 
recovery, it was estimated that, owing to an increase in sprout production, 
the stand would eventually be as dense or somewhat denser than before the 
fire. However, the sprouts were all small and growth rates indicate that full 
leaff head-size recovery would take at least 5 years longer and that many 
years would pass before the sprout stems were as tall as the old burned ones. 
Similar recovery of burned plants has been observed in Y. cam erosana and 
Y. rostrata. 

In Y. arizonica and Y. baccata the nodules commonly produce a terminal 
bud from which a rhizomatous extension or a stem develops (pi. 54). Al- 
though these stems are usually symmetrical and broadly conoid in shape, they 
are occasionally quite irregular. In Y. arizonica, such rhizomes are from 8 
to 15 cm. in diameter and 15 to 30 cm. long, while in Y. baccata they are 
approximately the same diameter but from 20 to 55 cm. long. The rhizomes 
are covered with a thick, checked bark and have typical dictyostele structure. 
They terminate in a very blunt bud that is covered with closely imbricated, 
large brittle scales. One or two adventitious sprouts develop along the 
rhizome's upper surface, and adventitious roots are produced along its under- 
surface. The roots are of a protostele structure. 

In Y. baccata the rhizomatous stems are apparently long-lived, and in many 
plants their continuous production and occasional branching forms a con- 
fused netlike system just beneath the soil surface. 

Benson and Darrow (3, p. 71) described Y. baccata as, "Stems 1 - several, 
subterranean or lying on the ground; leaf- rosettes clustered or solitary at 
ground level . . ." From this description it is not clear whether aerial pro- 
cumbent stems are of rhizomatous or of aerial origin. Since the stems of the 
two types of origin are quite distinct and result in different types of clumping, 
it is an important character in separating typical Y. baccata from its transitional 
forms. Rhizomatous, or basal sprout, stems never elongate after their terminal 
bud forms a leaf head. Although such stems are usually subterranean and 
horizonal, they are rather commonly aerial, or partially aerial, and may be 
ascending. In Y. baccata, such rhizomatous stems are easily distinguishable in 
that they are always short and thick, lack leaf bases or leaf scars, and are 
irregular in shape following leaf-head formation (pi. 59) . Procumbent stems 
of aerial origin are those stems that are produced by the growth and elongation 
of the leaf-head bud and stem. They are thus formed after the leaf head 
and are always aerial. Such stems, at least in Y. baccata, usually exhibit 
leaf remnants or leaf scars ; are generally long, slender, and uniform in shape ; 
and terminate in an erect head of leaves. In plate 59 the lower stem was 
only partly buried in the soil and the upper stem was entirely aerial. The 
lower stem is rhizomatous, and as it terminates in a leaf head the rhizomatous 
part of the stem lias completed its growth. The basal half of the upper 
stem is rhizomatous, whereas the top half is a typical assurgent-erect stem. 

Owing to the fact that rhizomatous stems are short and frequently from 
adventitious shoots, plants that are limited, or practically limited, to such stems 



YUCCAS OF THE SOUTHWEST 79 

are usually dense. Procumbent stems of leaf-head origin usually spread out 
from plant centers and result in a rather large open clump or large clumps with 
fairly open centers and dense marginal bands of upturned heads of leaves. In 
many cases the latter clumps become rather confusing by the formation of 
additional rhizomatous sprouts. Procumbent-stemmed plants of transitional 
forms of Y. baccata-Y. arizonica are rather common south of Vail, Ariz., and 
Y. baccata plants exhibiting aerial rhizomatous stems are rather abundant 
along the highway east from Cornudas, Hudspeth County, to El Paso 
County, Tex. 

In this group of plants, adventitious root production from aerial stems 
appears to be quite variable and may depend upon environmental conditions. 
In the vicinity of Valley Wells and Kessler Springs, Mojave Desert, Calif., 
many plants of Y . baccata variety vespertina and apparent hybrids between 
this variety and Y . schidigera were found, with aerial stems that lie on the 
ground. In several of these cases, the apical ends of the stems turn upward 
and extend fully a meter high. Where these stems come in contact with the 
soil, scattered roots occur. Occasionally the older, or lower, end of the stem 
has decayed completely through and roots are plentiful just below the up- 
turned apical end. Similiar plants of apparent Y. baccata-Y. arizonica hybrids 
occur south of Vail, Ariz. Many procumbent stems of these plants were 
examined and only occasionally were they rooted. 

In Y. brevifolia, the stem base remains bulb-shaped until the plants are 
between 2.4 to 4.0 m. tall, when from four to eight large rounded extensions, 
or protrusions, develop just above the roots. These protrusions are 5 to 8 cm. 
in diameter, 8 to 12 cm. long, and they are covered with a thick, relatively 
smooth whitish-gray bark. They extend obliquely downward and appear to 
be composed of comparatively active tissues. Except in forma herbertii the 
bases of old Y. brevifolia plants are elephant-foot-like and set flat on the 
ground. In these plants the protrusions are broad, fairly long, and flattened 
on their under side. They radiate from the center and are mainly dichoto- 
mously lobed (pi. 60). The roots are long, tough, and cordlike, and they are 
mainly limited to a marginal band immediately below the protrusion lobes. 

In several Y. brevifolia concentrations, approximately 5 percent of the plants 
produce basal sprouts. These sprouts are formed from nodules that develop 
slightly above the soil surface on the bases of the older stems. No nodules or 
sprouts have been found to produce roots, and only limited evidence indicating 
that the sprouts grow to an appreciable size has been obtained. 

Yucca brevifolia and its variety jaegeriana are mainly single-stemmed, but 
occasionally they have two or three stems and very rarely form a clump of 
several stems. Forma herbertii is a typical clumped plant. The clumps 
of herbertii are usually very large and so dense that they are impenetrable. 
Upon the excavation of parts of several clumps, it was found that all aerial 
stems were apparently attached to each other by rhizomatous stems and that 
the area and density of the clumps were still increasing. 

The rhizomatous stems appear to originate shortly following protrusion 
formation, from buds intermingled with, or occasionally above, the protrusion 
(pi. 61). These buds are small in diameter (10 to 15 mm.), elongated (15 
to 20 mm.), and sharply pointed. They are closely imbricated by large, 
triangular (altitude 12 mm.), brittle, brown scales. The rhizomes are hori- 
zontal or obliquely-ascending and range from 0.2 to 1.3 m. long and from 
1 to 3 cm. in diameter. When first produced, the rhizomes are unbranched, 



80 AGR. MONOGRAPH 17, U. S. DEPT. OF AGRICULTURE 

rather succulent, and covered with a thin, brown epidermis. The nodes are 
well marked by leaf scales, which are retained on the full length of the 
young rhizome. From the nodes unbranched, succulent roots develop. The 
rhizomes have a typical monocotyledonous dictyostele structure, whereas the 
roots exhibit a protostele structure. Apparently the young rhizomes develop 
rapidly and soon ascend to the soil surface, where they produce a head of 
leaves. 

Following sprout formation the young rhizomes lose their succulence and 
become woody and tough. During this transformation, the thin epidermis and 
bud scales are replaced by a thick, checked, dark-brown bark. Occasionally 
old rhizomes branch, and the branches produce sprouts close to the mother 
stems. 

In the plant shown in plate 61, the mother stem was approximately 2.0 m. 
tall. At the base of the stem are 8 protrusions, 10 rhizome buds, and 8 
rhizomes, which vary in length from 0.1 to 1.2 m. Of the 8 rhizomes, 6 have 
a thin epidermis and leaf scales at the nodes. The remaining 2 are covered 
with a thick, rough bark, and each is once-branched. 

Whether the formation of rhizomes in Y. brevifolia and variety jaegeriana 
is a normal behavior or a result of stem injury is uncertain. In the centers 
of several clumps of these forms a single decayed stem has been found. This 
evidence, in conjunction with the known sprout production following stem 
injury in other species, indicates that rhizomatous stem development in Y. 
brevifolia and its variety is in some way stimulated by shoot injury. Rhizo- 
matous stem development in this group may also depend upon the amount' 
of precipitation. It has been noted that in the lower, drier parts Y. brevifolia 
rarely forms more than one or two stems, but in the higher, moister areas 
two- and three-stemmed plants are quite common and occasionally small 
clumped plants are found. The annual precipitation in the type locality of 
forma herbertii is between 14 and 16 inches. 

Dehiscent-Fruited Species 

Yucca rostrata, Y. thompsoniana, and Y. whipplei and its varieties parishii 
(?) and caespitosa are the only capsular-fruited southwestern yuccas that do 
not, or rarely, form rhizomes. 

The leaf -head bases of Y. whipplei and variety parishii (?) are cup-shaped 
and extend 10 to 25 cm. into the soil. Except for a relatively small area in 
the center, these bases are covered with closely imbricated leaf bases. From 
the base centers, many sparsely branched roots extend into the soil. The 
plants never produce rhizomes or sprouts. 

The leaf -head base and roots of 1 -year-old seedlings of variety caespitosa 
are indistinguishable from those of the species or variety parishii ( ? ) . How- 
ever, when the seedlings of variety caespitosa are from 14 to 20 months old 
they develop from 4 to 16 buds in the axils of the outer leaves. Soon after 
the formation of these buds, the leaf -head base enlarges and somewhat flattens. 
This enlargement is followed by the production of succulent, fusiform roots 
that are similar to the roots of 5-month-old seedlings. The axillary leaf heads 
of many 20-month-old seedlings are very similar to younger mother leaf heads. 
However, they are somewhat flattened and their leaves are narrower than 
either those of the mother leaf head or of younger seedlings. 



YUCCAS OF THE SOUTHWEST »81 

Old field plants of variety caespitosa have a very small, thick, irregularly 
shaped, or lobed, caudex. This caudex is approximately 7 to 15 cm. long, 
5 to 9 cm. in diameter, and extends 5 to 10 cm. above the soil surface. The 
base of the caudex is extremely broad and bowl-shaped and extends very 
slightly into the soil. From the very bottom of the base, in an area hardly 
larger than that occupied by 20-month-old seedlings, long cordlike roots 
spread into the soil. The aerial part is composed of numerous axillary rosettes 
attached to the caudex. The production of axillary leaf heads is apparently 
repeated over and over until the resultant cluster is a very confused tangle ot 
crowded rosettes that completely covers the small caudex. 

In the apparent Y. whipplei-Y. whip pi ei var. caespitosa hybrids the rhi- 
zomes are initiated at the base of the stem, contiguous to the roots, in the form 
of small, elongated buds. Matured rhizomes are 2 to 4 cm. in diameter, 40 to 
95 cm. long, and covered with a comparatively thick bark. They produce 
adventitious roots and terminate in a bud, which ultimately forms a sprout. 
In several plants these rhizome sprouts have been found to produce axillary 
leaf heads. They are usually very close to the mother cluster, and unless the 
plant is carefully examined the original mother cluster and the rhizome- devel- 
oped cluster cannot be separated. From one to eight rhizomes have been found 
around the base of a single mother cluster of such apparent hybrids. 

Although the writer's studies of the subterranean system of Y. whip pi ei var. 
per cur sa have been limited, these studies, in conjunction with those of the 
aerial habit, indicate that sprout formation is mainly a result of rhizomatous 
development. 

The subterranean development of 2-year-old seedlings of Y. thompsoniana is 
similar to that of the majority of indehiscent-fruited species. They show no 
indication of rhizome development. However, young field plants, estimated 
to be 5 to 7 years old, exhibit a slightly swollen (2.5 to 4.0 cm.) leaf-covered 
base, from which a short, thick, rhizomatous stem extends straight down- 
ward for 12 to 20 cm. (see pi. 68,/i). This rhizomatous stem at the union 
with the aerial stem is 1.5 to 3.0 cm. in diameter, and it gradually increases 
toward the base to 3.5 to 6.5 cm. in diameter. At the latter point, it rapidly 
flares out to an irregularly shaped mass that is approximately 10 to 18 cm. 
in diameter and more or less flat on the bottom. From this swollen base, and 
rarely from the stemlike portion, numerous, elongated fusiform, succulent 
roots spread into the soil. The rhizome is covered with a comparatively 
rough, thick bark, whereas the roots are covered with a smooth, thin, dark-red 
epidermis. 

In older plants of Y. thompsoniana the massive rhizomatous base is very 
similar to the rounded base of Y. torreyi. It is from 20 to 30 cm. below 
the soil surface and is about 40 to 60 cm. in diameter (pi. 69). It is ex- 
tremely irregular in shape and exhibits many lobes and protrusions. The 
roots are long, thick, tough, and cordlike. Sprouts are rarely produced unless 
the old shoot is injured. 

In the preceding section it is shown that all other southwestern capsular- 
fruited yuccas form definite rhizomes (pis. 62 to 72). These rhizomes develop 
from buds formed on the axis, when the seedlings are between 5 and 9 
months old. In mature plants the rhizomes are of two types: Horizontal or 
somewhat oblique and vertical. 

Yucca rupicola, Y. reverchoni, Y. neomexicana, Y . gilbertnmu, Y . glauca, 
Y. bailey/, Y. angustissima, and Y. constricta are characterized by oblique or 



82* AGR. MONOGRAPH 17, U. S. DEPT. OF AGRICULTURE 

horizontal rhizomes (pis. 63, 66, and 67) that in the majority of cases follow 
a definite pattern of development. Shortly after the seedling rhizome is pro- 
duced, it gradually branches and rebranches (pi. 63). These branches spread 
out more or less horizontally in all directions. Following the development of 
these branches, or when the seedling is approximately 4 to 6 years old, short 
vertical branches, or risers^ are formed along the upper surface of the older 
rhizomes. Upon reaching the soil surface, each riser produces a head of 
leaves. Following leaf-head formation, the upper part of the rhizome, im- 
mediately below the young sprout, gradually increases in diameter and becomes 
very active. It soon produces several lateral rhizomes that branch freely. At 
the latter stage, the young shoot and its rhizome are similar to those of 4- to 
6-year-old seedlings. Soon after sprout formation, the old rhizome uniting 
the shoot with the mother plant appears to become less active (pi. 67). Al- 
though the connections usually remain alive in a dormant condition, occasionally 
they die and soon decay. 

The rhizome system of large plants is a very confused network that gen- 
erally lies from 10 to 60 cm. below the soil surface (pi. 66). Mature rhizomes 
are covered with a thick, rough bark. They lack nodes and leaf scales and 
terminate in blunt, scale-covered buds. Although the roots mainly occur 
immediately below the sprouts on the risers, they are also thinly scattered 
along the entire length of the horizontal rhizomes. 

The rhizomes of the various species differ considerably in size and number 
of sprouts they produce. Of the largest rhizomes found in several species, the 
range in length was from 15 to 30 cm. in Y. reverchoni to 75 to 300 cm. in 
a transitional form of Y. glauca; diameters ranged from 1.5 to 2.0 cm. in 
Y. reverchoni to 6.0 to 10.0 cm. in Y. bailey/. 

The dense, small clump habit of Y. reverchoni is mainly the result of 
rhizome-sprout production. Since the rhizomes are very short and highly 
branched (pi. 68,B), it is difficult to determine if the sprouts originate from 
risers or from the terminal buds of horizontal rhizomes. In several young 
plants, however, upturned buds of horizontal rhizomes strongly indicate that 
rhizome branches ultimately terminate in a leaf head. In this species, adventi- 
tious or axillary aerial leaf-head formation plays a secondary part in clump 
development. 

Although Y. gilbertiana is mainly a dense, compact clump, it is occasionally 
a rather open clump, with small scattered heads of leaves. In the former 
cases, the rhizomes are large in diameter, comparatively short, thinly branched, 
and lie obliquely in the soil. In such plants leaf-head production is mainly 
of aerial formation and rhizome-produced sprouts add very little to the 
density of the plants. Although plants of Y. gilbertiana exhibiting scattered 
heads of leaves have not been examined, they are undoubtedly of rhizomatous 
origin. 

The dense nature of Y. glauca and Y. baileyi is mainly the result of aerial 
leaf -head production. In these plants, as well as in Y. baileyi var. navajoa, 
there are usually one or two main rhizomes that are large in diameter, com- 
paratively short, and oblique. Rhizomatous sprouts are produced from short 
risers and from terminal buds of horizontal branches that develop around 
the upper parts of the main rhizomes. 

In Y. rupicola, Y. neomexicana, Y. angustissima, and many transitional 
forms of Y. glauca the heads of leaves are rather thinly scattered. In these 
plants secondary leaf heads are mainly of rhizomatous origin. The rhizomes 



YUCCAS OF THE SOUTHWEST 83 

are relatively thin in diameter and very long (pi. 67). They are mainly 
horizontal and, although much-branched, the branches are comparatively far 
apart. Although the sprouts are mainly produced from risers, they are occasion- 
ally formed at the ascending ends of long horizontal rhizomes. 

Large plants of Y. glauca are extremely difficult to destroy. In cleared and 
cultivated fields, regardless of deep seasonal plowing, it will continue to 
sprout for three or more years. Many plants that have been entirely deprived 
of shoots by fire or cutting put forth within a year or two as many or more 
shoots than they originally possessed. All of these sprouts or shoots develop 
from the netlike rhizome. 

Yucca elata is characterized by one or more large vertical rhizomes, which 
produce short risers. The formation of the rhizome system is very similar to 
that previously described in species exhibiting horizontal or oblique rhizomes. 
In Y. elata, however, the seedling rhizome continues to develop slowly down- 
ward throughout the plant's life (pis. 64, 65, and 70) or until it is replaced 
by another vertical rhizome of shoot origin. When the seedling is approxi- 
mately 5 to 6 years old, the large vertical rhizome usually produces one or 
more branches. These branches are always formed around the upper end of 
the rhizome or from 8 to 30 cm. below the soil surface. They are short and 
ascending and apparently always form secondary heads of leaves that are 
adjacent to, or very- close to, the mother seedling. During the development 
of the secondary leaf heads, the short, ascending rhizome from which it was 
derived increases in diameter and occasionally produces lateral buds. Appar- 
ently these buds either ascend to the surface and form additional leaf heads 
or, rarely, turn downward and produce a second larger vertical rhizome. 

The external morphology and root distribution of the rhizomes of Y. elata 
are similar to the descriptions previously given for species exhibiting horizontal 
rhizomes. The vertical rhizomes of mature plants are commonly 1.0 to 1.5 m. 
long and occasionally up to 2.0 m. Although they usually vary between 8 to 
15 cm. in diameter, several plants with rhizome diameters ranging between 
15 and 28 cm. have been excavated. The lateral or ascending branches vary 
from 15 to 20 cm. long and from 2.5 to 8.0 cm. in diameter. 

Yucca elata usually forms a rather tall, open clump that occupies a com- 
paratively small area. The clump is usually composed of one to five aerial, 
trunklike stems, with a few scattered shoots around their bases. Outside of 
the possibility of seedlings with twin shoots, all of the stems and shoots, 
except the seedling stem, are of rhizomatous origin. Aerial stem branching and 
leaf-head production add very little to the clump. Occasionally in sand dune 
areas the clumps are somewhat denser, larger in area, and shorter in height. 
Partial excavation of several of the latter plants indicated that their larger 
area and denser habit were undoubtedly caused by a higher frequency of 
vertical rhizome branching. The shorter height is probably an adaptation to 
windstorms and sandstorms. 

Both rhizomes and trunklike stems of Y. elata freely produce adventitious 
heads of leaves. Many plants with old leaf heads that had been completely 
destroyed by fire have been examined. In the less severely burned plants, the 
new leaf heads are formed in approximately equal numbers along the sides 
of the old aerial stems and from the rhizomes adjacent to the stems. All 
sprouts of severely burned plants are of rhizomatous origin and are close to 
the old burned stems. Occasionally along washes and ravines, the large ver- 
tical rhizomes of Y. elata are exposed. In these plants it is not uncommon to 



84 AGR. MONOGRAPH 17, U.S. DEPT. OF AGRICULTURE 

find the rhizomes replacing the aerial stems. Along the sides of the rhizomes, 
adventitious leaf heads are produced and the old leaf heads and aerial stems, 
as if too high, die and slough off. 

Many fallen or bent trunklike stems of Y. elata that were partially covered 
with soil and retained their original roots have been examined. The apical 
ends of these stems frequently grow upward for a meter or more. However, 
none of these stems have been found to produce either roots or heads of leaves 
along their sides. 

Undoubtedly one of the most interesting growth habits of southwestern 
yuccas is that of Y. elata in White Sands National Monument, N. Mex. The 
plants establish themselves in open spaces, between the dunes, and as the 
dunes move into these spaces, the plants gradually become covered. The stems 
of plants toward the edges of the rising dunes grow at a sufficient rate to 
keep their heads of leaves above, or nearly above, the dune surfaces, and the 
plants toward dune centers are completely covered. Consequently, toward the 
dune crests, the stems appear to become shorter and shorter until only the 
heads of leaves remain above the sand. Occasionally along the higher dune 
fringes only the flowering stalks or fruit stalks appear above the sand. 

As the dunes drift onward and the trunklike stems of Y. elata are grad- 
ually uncovered, numerous adventitious roots are revealed (pi. 71). These 
cordlike roots frequently extend 7 m. or more over the white sand. Although 
the old stems are mainly covered with a rough bark, occasional scattered 
bands of leaf remnants completely encircle them. Now and then rhizome 
risers are found adjacent to the trunklike stems. These risers are long and 
thin, and toward their ends the nodes are well marked by scales (pi. 72). They 
always appear to be of deep subterranean origin, and they are probably branches 
of the seedling rhizome. When the dunes have receded 2 or 3 m., the stems 
produce adventitious leaf heads toward their bases and the upper part of the 
old stem, as if it can no longer obtain nutriment or support itself, bends over 
and finally dies and sloughs off (pi. 72). Occasionally the ends of bent- 
over stems have the appearance of having been broken previously. These 
broken ends, in conjunction with parts of decayed stems close by, indicate 
that the height-adjustment process has probably occurred several times during 
the stem's existence. In several plants, the writer has estimated the stems 
originally to have been from 5 to 6 m. long. 

Rhizomes Produced in Yuccas 

In the preceding descriptions, it has been shown that the rhizomes of Yucca 
glauca, Y. angustissima, Y. neomexlcana, and the majority of other capsular- 
fruited yuccas are horizontal or oblique, whereas those of Y. elata are vertical. 
Although the rhizomes of these species have typical stem structure and ter- 
minate in a scale-covered bud, they are covered with a thick, rough bark and 
do not have nodes, internodes, and scales along their surface. The only 
species that produces rhizomes with scalelike leaves is Y. brevifolia. In this 
species, the rhizome apparently grows very fast and numerous brittle scalelike 
leaves are scattered along the young, long rhizome. Nodes and internodes are 
lacking or inconspicuous, and all of the older rhizomes are covered with a 
rough, thick bark. 

Within recent years, Holm (17) and others have suggested that the 
term "rhizome" should be restricted to subterranean stems, which represent 



YUCCAS OF THE SOUTHWEST 85 

* 
the primary axis developed directly from the plumule. It has been shown 
that in Y. glauca, Y. bailey/, Y. neomexicana, Y. elata and other capsular- 
fruited yuccas, the rhizomes are derived directly from the primary axis when 
the seedlings are quite young. In Y. brevifolia, however, they are produced 
from active tissue at the base of the aerial stem when the plant is rather old. 
In the fleshy-fruited yuccas, rhizomes are quite rare. Although they occasion- 
ally occur in Y. baccata and Y. arhonka, and possibly Y. torreyi and Y. 
schidigera, they are hardly more than a modified basal sprout. Such rhizomes 
are short, rarely more than 25 to 30 cm. long, stout, covered with a thick, 
rough bark, and terminate in a very large, blunt, scale-covered bud. They 
occur only on large mature plants. In all fleshy-fruited yuccas, basal sprouts 
are mainly derived from nodulelike growths, produced in active tissues at 
the base of the aerial stems. Such nodules lack a terminal bud and never 
exhibit any lateral growth. 

Both the typical rhizomes of the capsular-fruited yuccas and the modified 
rhizomes of the fleshy-fruited yuccas produce scattered, aerial shoots and 
adventitious roots, and the nodules develop into a single basal sprout. Under 
the section "Transplanting and Asexual Propagation," it is shown that the 
nodules are useful in vegetative propagation, whereas the rhizomes appear of 
little or no value in such work. 

Rhizomatous yuccas are characteristic of the more humid regions, but the 
fibrous-rooted plants are mainly limited to the drier regions. In the Y. rupicola 
alliance, Y. rostrata is strictly a fibrous-rooted species and occurs in the very 
dry parts of southwestern Texas. On the other hand, Y. rupicola has a well- 
developed rhizome and occurs in the humid xerophytic-mesophytic transitional 
zones in west-central Texas. Between these species is Y. thompsoniana, with 
a poorly developed rhizome growing in a slightly more humid climate than 
that of Y. rostrata, and Y. reverchoni, with a fairly well-developed rhizome 
growing in a slightly less humid area than that of Y. rupicola. Y. whipplei 
and its varieties caespitosa and parishii (?) are found in southern California, 
whereas variety percursa extends along the coast in south-central California. 
Y. whipplei var. percursa is rhizomatous and subjected to considerably more 
rainfall than Y. whipplei and its varieties caespitosa and parishii (?), which 
rarely produce rhizomes. Similar correlations between the amount of rainfall 
and the development of rhizomes occurs in Y. brevifolia, and, as a general 
rule, it applies to all species and forms. 

RECOVERY OF HARVESTED YUCCAS 

Near Kingman, Ariz., in May 1944, all sprouts of five plants of Yucca 
schidigera were removed and the trunks cut at various heights from soil level. 
In June 1945 the number of sprouts these plants had produced since cutting 
were from zero to nine, and these sprouts were from 8 to 23 cm. long. 

Although the results concerning recovery of sprout formation were based 
on insufficient material, they indicate that the greater the injury to the shoot, 
the less likely the plant is to recover. Plants with only the sprouts and leaf 
heads removed formed many more sprouts than those plants that had stems 
cut as well as sprouts and leaf heads removed. In addition to having a larger 
recovery of sprouts, the sprouts of less damaged plants appear to be more 
vigorous. 



86 AGR. MONOGRAPH 17, U. S. DEPT. OF AGRICULTURE 

On the United States Department of Agriculture Field Station grounds, at 
Tucumcari, N. Mex., 20 plants of Y. glauca (hybrids?) were cut in July 1942. 
The first 10 of these plants were cut just below the leaf head at ground level, 
and the second 10 cut 15 cm. below the soil surface. In October 1942 and 
June 1943, respectively, 5 and 2 more plants were cut at ground level. 

Of 574 leaf heads cut from the 20 plants there were 553 sprouts after 
3 years; of the 142 leaf heads cut from the 5 plants, 170 sprouts had appeared 
31 months later; and of the original 52 leaf heads cut from the 2 plants, 73 
sprouts appeared at the end of the year. 

Thus, it is shown that harvested leaf heads of Y. glauca are rapidly replaced, 
and that frequently more heads of leaves are formed than the plant originally 
possessed. Although in the majority of cases the original number of leaf heads 
is not recovered until approximately 3 years after cutting, in many cases it 
is recovered within 3 months to a year. Five of the twenty plants involved 
in the experiment formed at 5]/ 2 months after cutting as many or more leaf 
heads than they originally had, and 8 of the 20 plants had more than their 
original number in 1 year's time. 

Regardless of rapidity of recovery in the number of leaf heads, weight 
recovery of the leaf heads is comparatively slow. Within the 3-year period, 
only 54.61 percent of the original leaf-head weight was recovered in plants 
cut at ground level, and only 43.15 percent of the weight for those plants cut 
below soil level. The data indicate that total weight recovery would require 
approximately 5y 2 years in the former group of plants and 6% years in the 
latter group. However, because weight and recovery relationships of plants 
cut at soil level were based upon more leaf heads than the plant originally 
possessed, it is very likely to take considerably longer than 5^2 years before 
these plants recover leaf heads as large as those that were cut. 

There is very little evidence indicating that seasonal injury or weather con- 
ditions affect the amount or rapidity of recovery. The low percentage of 
recovery immediately following the October 1942 cutting indicates that plants 
recover less rapidly if injured late in the year. On the other hand, for 6 
months following this cutting precipitation was only 0.89 inch, whereas it 
is normally 4.24 inches for the period. That short droughts do not affect 
recover}' is made evident by the large growth following the June 1943 cutting. 
This cutting was made in the midst of an 11 -month hot, dry spell, when the 
rainfall was approximately three-fifths normal. Total precipitation for the 
3-year experimental period was 41.37 inches, compared with a long-time 
normal 3-year precipitation average of 49.50 inches. 

In the preceding experiments, one plant formed a single inflorescence when 
the recovered sprout was 23 months old and another plant produced two 
inflorescences when the leaf heads were 36 months old. 

Near Tayler Well, United States Jornada Experimental Range, Las Cruces, 
N. Mex., in June 1943, all heads of 40 plants of Y. el at a were cut. Four months 
after harvesting, 9 of the plants showed no recover} 7 . The remaining plants, 
however, had produced 129 leaf heads, whereas a total of only 97 was 
originally cut. Approximately 67 percent of the new leaf heads were of 
rhizomatous origin, and the remaining 33 percent were produced either at 
the apices of the cut stems or scattered along the sides of these stems. Two 
years after harvesting the plants had 108 leaf heads, about half of which were 
of rhizomatous origin and half of aerial stem origin. Forty-three percent of 



YUCCAS OF THE SOUTHWEST 87 

the aerial-produced leaf heads originated at the apices of the cut stems and 
57 percent along the sides of these stems. 

On the same range, near West Well, in 1942, nine plants were harvested. 
All of the aerial growth of six of these plants was cut at soil level, and all 
leaf heads of the remaining three were cut immediately below the fresh leaves. 
Approximately 32 months after harvesting the plants with all aerial growth 
removed showed about 83 percent recovery in number of leaf heads, and the 
plants cut immediately below the leaves showed 90 percent recovery. 

In these experiments all growing points of trunklike stems were destroyed. 
Although many of these stems died, it has been indicated that others formed 
new sprouts. Campbell and Keller (8) reported that all stems with growing 
points that have been destroyed soon die. 

Although the experiments were abandoned before completion, it was esti- 
mated that recovery of leaf heads to full size would require at least 5 to 6 
years and that harvested-weight renewal could not be accomplished unless 
rodent control measures were applied. 

Many southwestern range foresters and botanists have been consulted by the 
writer relative to Y. elata's recovery following harvesting. Although the major- 
ity of these scientists believe that Y. el at a will recover following shoot injury 
or harvesting, several are of the opinion that complete recover)- is improbable 
and that continued harvesting would soon kill the plants. An eminent Arizona 
ecologist (Forrest Shreve) found that following the Y. el at a harvesting during 
the first World War many plants failed to recover and that only 5- to 10- 
percent recovery occurred in a large area near Orogrande, N. Mex. The 
latter failure was mainly attributed to rodent damage and to shifting sand, 
which killed or covered the developing shoots. 

GROWTH RATES AND AGE OF YUCCAS 

Campbell and Keller (8 ) reported that the leaves of Yucca el at a grew rap- 
idly during the first year and usually died at the end of the third or fourth 
year. In the nursery of the writer, the outer leaves of 4-year-old seedlings of 
Y. elata, Y . glauca, Y. baileyi, and Y. constrkta die when the plants are ap- 
proximately 4 years old. The leaf heads of these plants are about equal in 
size to those of the parent seed plant. The leaf heads of Y. neomexicana 
seedlings reach such maturity in 3 years. On the other hand, the leaf heads of 
3-year-old seedlings of Y. rupicola, Y. revercboni, and Y. tbompsoniana are 
about one-half average size, and it appears that they would not be so large 
as those of their parental seed plant until they are at least 5 years old. 

Leaf-head development of the fleshy-fruited yuccas is considerably slower 
than that of the capsular- fruited species. In the former group, the leaf heads 
of 414-year-old seedlings of Y. torreyi, Y. arizonka, and Y. baccata are ap- 
proximately one-half parental size and all leaves are considerably smaller 
than those of the parents. The leaf heads of 3-year-old Y. schidigera and Y. 
scbottii seedlings are between one-fourth and one-half parental size, while 
those of 3-year-old Y. faxoniana seedlings are quite small and less than one- 
fourth the size of those of the parent seed plant. It is apparent in these 
species that no leaf heads will be as large as those of the parent seed plant 
until they are 5 to 7 years or more old. 

As the longevity of fresh leaves and, consequently, the size of leaf heads 
largely depend upon the amount of water the plant received, undoubtedly the 



88 AGR. MONOGRAPH 17, U. S. DEPT. OF AGRICULTURE 

age of flowering is a more accurate index of leaf-head maturity. In the 
writer's nursery, the majority of seedlings of practically all southwestern yuccas 
are from 5 to 6 years old. The species that have flowered and the age at 
which these species flowered are as follows: 

Species Flowering age 

Yucca rupicola 3 years and 8 months 

Y. reverchom 4 years and 10 months 

Y. neomexicana Do. 

Y. constricta Do. 

Y. thompsoniana 5 years and 4 months 

Y. glauca (hybrid?) 5 years and 1 month 

Y. glauca (hybrid?) 6 years and 1 month 

Y. arizonica Do. 

Y. elata Do. 

In all of the preceding species, only one or two of several plants flowered. 
At the time of flowering the leaf heads of Y. elata, Y. glauca, Y. constricta, 
and Y. neomexicana were slightly larger than those of average native plants, 
whereas the rest were about the same size as those in flowering field plants. In 
the first Y. glauca (hybrid?) two secondary leaf heads were formed shortly 
after flowering. Both of these secondary leaf heads flowered 10 months after 
their formation. 

In March 1928 numerous seedlings of Y. ivbipplei from seeds collected in 
Claymine Canyon, Santa Ana Mountains, Calif., were set out in the Rancho 
Santa Ana Botanic Garden, Orange County. In 1934, when the seedlings were 
approximately 7 years old, Wolf (33) found that 7 of the 77 plants that were 
still alive flowered. 

For Y. elata, Campbell and Keller (8) reported an average annual stem 
growth of 1.07 inches (2.55 cm.), with a maximum growth of 2 inches 
(5.0 cm.) in favorable rainy years and a minimum growth of 0.5 inch (1.3 cm.) 
in a dry year. In the Riverside, Calif., nursery stem growth rates are shown 
in table 3. 

These data show that Y. brevifolia and its variety jaegeriana are the fastest 
growing southwestern yuccas. The annual growth of the stems of the variety 
ranged from 8.12 to 16.25 cm., with an average of 11.74 cm., whereas that 
of the species varied from 5.22 to 8.12 cm., with an average of 6.64 cm. 
The writer has obtained accurate records of two older plants of Y. brevi- 
folia growing in Riverside, Calif. These plants were subjected to minimum 
irrigation and were in well-drained disintegrated granite soil. When the plants 
were 21 years old, their stems were unbranched and 109 and 141 cm. tall, 
respectively. These plants exhibit an annual growth of 5.20 and 6.74 cm., 
respectively, with an average of 5.94 cm. McKelvey (24 ) pointed out that 
information and publications of Ernest Braunton indicated that a native plant 
of Y. brevifolia grew 1 m. in 6 or 7 years. 

Approximately 6-year-old seedlings of Y. elata, derived from plants in 
Vail Junction, Ariz., exhibit annual stem growths of 1.50 and 2.03 cm., and 
those from State College, N. Mex., show annual stem growth 1.68 to 2.20 
cm. The average annual stem growth of these seedlings is 1.78 and 1.91 cm., 
respectively. These growth rates are somewhat lower than those reported by 
Campbell and Keller (8). On the other hand, Y. elata seedlings from a plant 
in White Sands National Monument, N. Mex., exhibit an annual growth of 
4.68 to 9.12 cm., with an average of 6.83 cm. The stems of these plants 
grew approximately 2.6 times as fast as those reported by Campbell and 



YUCCAS OF THE SOUTHWEST 89 

Keller and about 3.7 times as fast as the stems of Y. elata seedlings at Vail 
Junction and State College. 

Yucca scbottii appears to be the fastest growing southwestern fleshy-fruited 
yucca. Approximately 5 -year-old seedlings of three introductions of this 
species exhibited an annual growth range of 2.78 to 4.28 cm., with an 
average of 3.35 cm. 

The Rancho Santa Ana Botanic Garden, Orange County, Calif., has 116 
seedlings of Y. schidigera on a southern hillside slope of heavy gravelly loam. 
These plants were grown from seed collected at Tecate, San Diego County, 
Calif. They have not been irrigated since the young seedlings were planted 
on the hillside. On March 30, 1943, when the plants were approximately 
151/2 y ears old, the writer found 63 of the plants flowering. The heights 
of the stems of the 116 plants ranged from 0.0 cm. to 100 cm., with 30 
plants having an average height of 40 cm. The annual average growth of 
the stems is 2.65 cm. Although none of the plants had more than one main 
shoot, young sprouts around the base of the old stems ranged from to 7 
in number, and averaged 1.62 cm. in height. 

Aside from the preceding unirrigated plants, several seedlings were planted 
on a hilltop in more sandy soil and subjected to semi-irrigation. In the 15^2 
years these plants had made exceptionally good growth. The only one 
measured had four main stems, each of which had several leaf heads around 
the top, with one to three inflorescences. The four stems ranged from 40 
to 120 cm. tall. The average annual growth of the tallest stem is 7.7 cm. 

Campbell and Keller (8) concluded that growth in Y. el at a largely depends 
upon the amount of rainfall and that practically all stem growth is made in the 
summer. Although the data from the Rancho Santa Ana Botanic Garden 
plants and from the writer's nursery seedlings confirm these conclusions, the 
variations in growth rates of individual introductions indicate that growth 
rates also depend upon the genetical constitution of the plants. 

Very little can be said relative to the age of yuccas. McKelvey (24) cited 
Hollick (16, p. 249) who in reference to Y. brevi folia wrote, "The ages of 
certain individual trees have been estimated at approximately 600-800 years. 
The age of the largest one known, located about eighteen miles east of Lan- 
caster, in Antelope Valley, was estimated at 1,000 years or more. It was eighty 
feet in height and nine feet in circumference." McKelvey also cited Sudworth 
(30, p. 203), who in reference to Y. schidigera stated, "No definite statement 
can be made concerning the age limit of this yucca, which, however, can hardly 
be less long-lived than the Joshua tree. Messrs. C. R. Orcutt and S. B. Parish, 
who know the tree yuccas from long observation, both inform the writer 
[Sudworth] that the Mohave yucca is an exceedingly persistent but very 
slow grower in its native habitat, scarcely any change having been perceived 
in trees under observation for the last twenty-five years." 

It is the writer's opinion that the majority of yuccas are extremely long- 
lived, and that plants of several species are as old or older than those of 
Y. brevifolia. It has been shown that Y. faxoniana, Y. schidigera, and Y. 
torreyi occasionally have stems 6.5, 2.5, and 4.3 m. tall, respectively, and that 
3- or 4-year-old seedlings of these species are quite small and exhibit no 
stems. Y. arizomca occasionally produces stems 2.4 m. tall. The annual 
growth rate of seedling stems in this species was shown to be 0.27 cm. Based 
upon these measurements, the older stems of Y. arizomca would be approx- 
imately 880 years old. Such plants, however, are undoubtedly considerably 



90 AGR. MONOGRAPH 17, U.S. DEPT. OF AGRICULTURE 

older, as the growing conditions of native plants are poorer than those upon 
which the growth rate is based and no allowance has been made for secondary 
leaf-head formation and stem branching. 

Although the majority of rhizomatous species do not form tall stems, they 
often produce large clumps. Such clumps frequently cover an area of 10 
to 20 square meters, and clumps covering 40 square meters or more are not 
uncommon in the Y. glauca alliance. Morphological studies of many of 
these clumps and of seedlings derived from several of them indicate that the 
majority, if not all clumps, are either a simple plant or a clone. The growth 
rates and slow development of clumps in seedlings, in conjunction with the 
large size and extreme persistence of native plants, lead the writer to believe 
that many of the clumps of the rhizomatous species are as old as or older than 
any of the stemmed yuccas. 

The longevity of yuccas has undoubtedly contributed considerably to their 
diversity. Variations produced are preserved for many years, and, hence, their 
chances of spreading, intermingling, and recombining are probably greatly 
augmented. 

TRANSPLANTING AND ASEXUAL PROPAGATION 

While developing a new nursery in May 1946, the writer transplanted 2- 
to 3-year-old seedlings of practically all southwestern yucca species. The 
seedlings were simply removed bare-rooted to the new location and immediately 
planted and watered. All species having a fibrous root system continued to 
grow and exhibited no ill effects from the transplanting. Several of the plants 
having a rhizomatous system, however, showed a definite set-back in that 
the outer leaves of the leaf head died. Out of some 130 seedlings transplanted, 
only 2 finally died. These 2 plants were of Yucca elata. and upon examining 
them it was found that part of the rhizome of each of them had been cut off. 

In the past few years, several 5- to 6-year-old seedlings of practically all 
species have been transplanted without a loss. These plants were transplanted 
at various times through the year and handled in the same manner as the 
preceding ones. In the rhizomatous group and Y. brevi folia, approximately 
35 percent of the transplants exhibited dead leaves within a week and from 
one-fourth to one-third of their outer leaves finally died before the plants re- 
established themselves. In several cases apparently excess watering during the 
transplanting caused the outer leaves of the imbricated bud to become dis- 
eased. These leaves showed a brown band near their bases and, in a few cases, 
sloughed off. Such plants, however, rapidly recovered upon discontinuance of 
irrigation. 

Young seedlings of Y. neomexicana, Y. gilbertiana, Y. whipplei, and Y. 
baileyi var. navajoa have been successfully transplanted from native concen- 
trations to the writer's nursery. These seedlings were carefully dug and shipped 
in slightly moist paper. Two to three weeks elapsed between digging and 
planting. 

Along the highways in southwestern Texas are many beautiful plants of 
Y. torreyi, Y. treculeana, and Y. faxoniana, which were transplanted from 
native concentrations. According to reliable sources the plants were given 
very little care following transplanting, and practically all of them survived. 
They are fibrous-rooted species and were from 3 to 8 feet tall when moved. 
The writer has been informed that a few plants of Y. elata were also sue- 



YUCCAS OF THE SOUTHWEST 91 

cessfully transplanted along the highways. The transplanting of this species, 
however, was found to be very expensive, as the subterranean system frequently 
extended 4 feet into the ground and contained very few side, or feeder roots. 
According to the information supplied, Y. elata is not hard to transplant, pro- 
vided the entire root [rhizome] is dug up. 

Yucca baccata, Y. torreyi, and Y. schidigera are easily propagated by basal 
nodules and young sprouts. The sprouts are simply cut off close to the 
mother stem, preferably with a few roots; the cut side allowed to dry a few 
days; and planted. The growth of these cuttings appears to be normal and 
uninterrupted. On the other hand the writer has had very little success in 
asexual propagation of rhizomatous species. Many sprouts of Y. elata, Y. 
glauca, and Y. neomexicana have been carefully dug, cut, and planted. These 
sprouts frequently possessed rhizomes 0.5 to 1.0 m. long, which were well 
supplied with fibrous roots. Following planting, these plants were subjected 
to fairly heavy, light, or no irrigation. All sprouts receiving heavy irrigation 
died within 4 weeks after planting, whereas the majority of those having 
light or no irrigation died within 6 to 8 weeks. Only 1 sprout of Y. neo- 
mexicana and 2 of Y. glauca finally grew. All were from the lightly 
irrigated plot, where originally some 12 sprouts had been planted. 

In testing the possibility of vegetative propagation in the rhizomatous species 
the writer selected in native concentrations three plants each of Y. elata, Y. 
neomexicana, and Y. glauca. The sprouts of each plant were separated from 
the mother stem or another sprout by a rhizome at least 50 cm. long. In each 
case a narrow trench was dug between the sprout and the mother stem and a 
5- to 10-cm. section was removed from the rhizome. Thus, the sprout was 
Completely separated from the mother stem. During the following year the 
plants were checked several times. None of the sprouts showed any effect 
of the separation and all of them appeared to continue normal growth. It 
thus appears that the losses in vegetative propagation of the rhizomatous 
species are mainly caused by the method of digging the sprout. The roots of 
such sprouts usually occur immediately below the sprout's leaf head, and it 
is very likely that too many of them are cut during the digging. 



92 



AGR. MONOGRAPH 17, U. S. DEPT. OF AGRICULTURE 



Table 3. — Stem growth rates of some yucca species 1 grown at the 
Riverside, Calif., nursery 



Seedling and source 



Age 



Average 
annual 
growth 



Annual 
growth 



Height 

of 
stem 



Y. arizonica: 
From Nogales, Ariz, 



Y. schottii: 
From Nogales and Patagonia, Ariz. 



Y. elata: 

From Vail Junction, Ariz 

From State College, N. Mex 

From White Sands National Mon- 
ument, N. Mex 

Y. brevifolia: 
From Moronga Valley, Calif 

Y. brevifolia var. jaegeriana: 
From Pierce Ferry Road, Mohave 
County, Ariz 



Months 
*71 



56 



71 



71 



71 



62 



62 



Cm. 
0.27 



3.35 



1.78 



1.91 



6.83 



6.64 



11.74 



Cm. 
1.34 

to 



2.78 
3.00 
3.85 
4.07 

4.28 

1.50 
2.03 

,'1.68 
2.03 
;2.20 



r 5.22 
6.58 
,8.12 

8.12 
10.84 
16.25 



Cm. 



5 13 
e 14 

18 
19 
20 



12 

10 
12 
13 

28 
31 
49 
54 

27 
34 

42 

42 
56 

84 



i Equally as old seedlings of other stemmed species (Y. faxoniana, Y. torreyi, 
Y. schidigera, Y. thornberi ?, and Y. thompsoniana) exhibit no aerial stems. 

2 71 -month-old seedlings germinated April 10, 1943; others, except last six of Y. 
schottii, germinated Jan. 21, 1944. The six Y. schottii seedlings germinated July 10, 
1944. 

8 At time of measurement. 

4 4 plants. 

5 2 plants. 

6 3 plants. 



YUCCAS OF THE SOUTHWEST 93 

LITERATURE CITED 

(1) Anonymous. 

1885. THE germination OF cyclamen and yucca. Gard. Chron. 24: 216. 

(2) Bell, W. H., and Castetter, E. F. 

1941. ETHNOBIOLOGICAL STUDIES IN THE AMERICAN SOUTHWEST. VII. THE 
UTILIZATION OF YUCCA, SOTOL, AND BEARGRASS BY THE ABORIGINES IN 

the American southwest. N. Mex. Univ. Bui. 372, Biol. Ser. 5, 74 pp. 

( 3 ) Benson, L., and D arrow, R. A. 

1944. A MANUAL OF SOUTHWESTERN DESERT TREES AND SHRUBS. Ariz. Univ. 

Bul. 15 (2) (Biol. Sci. Bui. 6), 411 pp., illus. 

(4) [Botkin, C. W.] 

1944. THE FIBER AND SAPONIN OF DIFFERENT SPECIES OF YUCCA. N. Mex. 
Agr. Expt. Sta. Ann. Rpt. 55: 51—54, illus. 

(5) and Shires, L. B. 

1944. tensile strength of yucca fibers. N. Mex. Agr. Expt. Sta. Bul. 316, 
29 pp., illus. 

(6) Shires, L. B. and Smith, E. C. 

1943. fiber of native plants in new Mexico. N. Mex. Agr. Expt. Sta. Bul. 
300, 38 pp., illus. 

(7) Campbell, R. S. 

1929- A CASE OF PHYLLODY IN YUCCA EL ATA. Bot. Gaz. 88: 109-1 10, illus. 

(8) and Keller, J. G. 

1932. GROWTH AND REPRODUCTION OF YUCCA ELATA. Ecology 13: 364-374, 

illus. 

(9) Cory, V. L. 

1930. A NEW SALVIASTRUM FROM THE EDWARDS PLATEAU OF TEXAS. Rhodora 

32: 89-90, illus. 

(10) Coville, F. V. 

1893. botany OF the death valley expedition. U. S. Natl. Herbarium Con- 
trib. 4: 203. 

(11) Cruse, R. R. 

1949. A CHEMURGIC SURVEY OF THE DESERT FLORA IN THE AMERICAN SOUTH- 
WEST. Econ. Bot. 3: 111-131, illus. 

(12) Engelmann, G. 

1878. NOTES ON the genus yucca. Acad. Sci. St. Louis, Trans. (1873) 
3: 17-54. 

(13) 

1881. SOME ADDITIONS TO THE NORTH AMERICAN FLORA. Bot. Gaz. 6: 223- 

225. 

(14) FORSLING, C. L. 

1919. CHOPPED SOAPWEED AS AN EMERGENCY FEED FOR CATTLE ON SOUTH- 
WESTERN ranges. U. S. Dept. Agr., Dept. Bul. 745, 20 pp., illus. 

(15) Haines, L. 

1941. variations IN yucca whipplei. Madrona 6: 33-45, illus. 

(16) Hollick, A. 

1932. deserts of the southwest. N. Y. Bot. Gard. Jour. 33: 247-250. 

(17) Holm, T. 

1929. THE APPLICATION OF THE TERM "RHIZOME." Rhodora 31: 6-17. 

(18) Jones, J. M., and Conner, A. B. 

1918. THE UTILIZATION OF YUCCA FOR THE MAINTENANCE OF CATTLE. Tex. 

Agr. Expt. Sta. Bul. 240, 23 pp., illus. 

(19) Katz, A., Hall, A. G., and Petersen, R. 

1947. LIGNIN AND VANILLIN FROM YUCCA BREVIFOLIA OF CALIFORNIA, hi Pro- 
ceedings of Conference of Cultivated Drug and Associated Economic Plants 
in California. Calif. State Dept. Ed. and Calif. Polytech. School, pp. 
85—109, illus. 

(20) Kearney, T. H., and Peebles, R. H. 

1942. FLOWERING PLANTS AND FERNS OF ARIZONA. U. S. Dept. Agr MlSC. 

Pub. 423, 1069 pp., illus. 

(21) Lapham, M. H. 

1948. THE desert storehouse. Sci. Monthly 66: 451-460, illus. 



94 AGR. MONOGRAPH 17, U. S. DEPT. OF AGRICULTURE 

(22) Laudermilk, J. D., and Munz, P. A. 

1935. PLANTS IN THE DUNG OF NOTHROTHERIUM FROM GYPSUM CAVE, NEVADA. 

Carnegie Inst. Wash., Paleontol. Contrib. 453: [29]-37, illus. 

(23) and Munz, P. A. 

1938. PLANTS IN THE DUNG OF NOTHROTHERIUM FROM RAMPART AND MUAV 

caves, ARIZONA. Carnegie Inst. Wash., Paleontol. Contrib. 487: [271]- 
281, illus. 

(24) McKelvey, S. D. 

1938. YUCCAS OF THE SOUTHWESTERN UNITED STATES: PART ONE. 150 pp., 

illus. Jamaica Plain, Mass. 

(25) 

1947. YUCCAS OF THE SOUTHWESTERN UNITED STATES: PART TWO. 192 pp., 

illus. Jamaica Plain, Mass. 

(26) Merriam, C. H. 

1893. NOTES ON THE GEOGRAPHIC AND VERTICAL DISTRIBUTION OF CACTUSES, 

yuccas, and agave . . .N. Amer. Fauna 7 (2): 345-359. 

(27) Munz, P. A. 

1948. let's save the short-leaved joshua tree. Natl. Parks Mag. 22 
(92): 8-12, illus. 

(28) Rydberg, P. A. 

1922. FLORA OF THE ROCKY MOUNTAINS AND ADJACENT PLAINS. Ed. 2. 1143 

pp. New York. 

(29) Standley, P. S. 

1920. trees and shrubs OF MEXICO, part one. U. S. Natl. Herbarium 
^Contrib. 23, 169 pp. 

(30) Sudworth, G. B. 

1908. forest trees of the pacific slope. 441 pp., illus. Washington, D. C. 

(31) Trelease, W. 

1902. the yucceae. Mo. Bot. Gard. Ann. Rpt. 13: 27-133, illus. 

(32) 

1907. additions to the genus yucca. Mo. Bot. Gard. Ann. Rpt. 18: 225- 
230, illus. 

(33) Wolf, C B. 

1935. California plant notes I. Rancho Santa Ana Bot. Gard. Occas. Papers 
Ser. 1 (1): 31-43. 

(34) 

1943. yucca schidigera, mo j ave yucca. Rancho Santa Ana Bot. Gard. Pop. 

Inform. Leaflet. 57, [4] pp. 

(35 

1945. the JOSHUA TREE. In California Wild Tree Crops. Rancho Santa Ana 

Bot. Gard. Unnumb. Pub., pp. [5]-17, illus. 

(36) Wooton, E. O. 

19 18. CERTAIN DESERT PLANTS AS EMERGENCY STOCK FEED. U. S. Dept. Agr. 

Dept. Bui. 728, 31 pp., illus. 

(37) Yanovsky, E. 

1936. FOOD PLANTS OF THE NORTH AMERICAN INDIANS. U. S. Dept. Agr. 

Misc. Pub. 237, 84 pp. 



INDEX TO YUCCA SPECIES 

[Synonyms are in italic type; page numbers of principal entries are 
set in heavy-face type] 



Page 

Alpine yucca 49 

Baccatae 24, 29, 30, 50 

Banana yucca 27 

Barreta, palm 18 

Beargrass 47 

Big Bend yucca 38 

Blue-form yucca 22, 23-25, 29, 73 

Cactus, sword 20 

Cactus-yucca 30 

Chaenocarpa 13, 16, 55, 70 | 

Chaparral yucca 32 

Clistocarpa 16, 67 

Clistoyucca arborescens 31 

brerijolia 31 

Constrictae 55 

Datil 27 

De datils 21 

Dwarf forms 50, 53, 60 

Elatae 55 

Glaucae 55 

Great Plains yucca 47 

Hesperoyucca 16, 70 

ivhipplei 33 

Hoary yucca 20 

Hybrids, apparent 18, 

20-25, 29, 30, 35, 36, 40, 42, 
43, 27, 52, 56-59, 62-65, 70, 73-75, 
81; pis. 14, 28, 29, B, C, 31-33, 
36, D-F, 37, C-F, 38-43, 48, 50. 

Joshua-tree 11, 30 

Mohave yucca 24 

Moth, yucca 12, 13, 65-69 

Mountain yucca 20 

New Mexico yucca 43 

Our-Lord's Candle 32 

Palma 19 

Palm barreta 18 

Palmella 59 

Palm-pita 21 

Palm samandoca 18 

Quixote yucca 32 

Salt Lake Desert yucca 44 

Samandoca, palm 18 

Samuela carnerosana 18 

faxoniana 19 

San Angelo yucca 41 

Sarcocarpa 16, 69 

Soaptree 59 

Soapweed 47, 59 

Spanish-bayonet 21 

Spanish-dagger 21 

Sword cactus 20 

Torrey yucca 20 

Trans-Pecos yucca 39 

Treculeanae 24 

Tree-yucca 30 

Yucca, alpine 49 

angustijolia 47 



Page 
Yucca angustijolia — Con. 

var. elata 59, 60 

var. radiosa 59 

angustissima 3, 

15, 17, 45, 51, 52, 53, 54-58, 
62-65, 70, 71, 76, 81, 82, 84; pis. 
34, C, 35, C, 46. 

dwarf form 53 

arborescens 30 

argospatha 21 

arizonica 3, 

6, 7, 9, 17, 20, 26, 27, 28-30, 66, 
68, 71, 73, 76, 77, 78, 79, 85, 
87-89, 92; pis. 10, 54, B. 

arkansana 54 

aspera 21 

australis 18, 19 

baccata 3, 

4, 6, 8-10, 19, 20, 23, 24, 27-30, 
37, 40, 44, 45, 50, 59, 71, 73-77, 
78, 79, 85, 87, 91; pis. 11, 54, 
A. 59. 

var. breiifolia 26, 29, 30 

var. macrocarpa 22, 24 

var. vespertina 25, 

27, 28, 29, 75, 79; pis. 12, 13. 

baileyi 3, 

15, 17, 44, 48, 49-52, 53-58, 62, 
71, 81, 82, 85, 87; pis. 34,B, 35,B, 
36,5, 44, 62. 

dwarf form 50 

var. navajoa 49, 

51, 52, 57, 72, 73, 82, 90; pi. 45. 

banana 27 

Big Bend 38 

brevifolia 2, 

3, 7, 8, 10, 17, 25, 26, 28, 30-32, 
37, 67, 70-73, 77, 79, 80, 84, 85, 
88-90, 92; pi. 15. 

f. herbertti 30, 

31, 32, 79, 80; pis. 18, 61. 

var. jaegeriana 7, 

11, 30, 31, 32, 73, 79, 80, 88, 92; 
pis. 16, 17, 53,B, 60. 

var. uolfei 31 

calijornica 24, 25, 33 

campestris 62, 64, 74 

canaliculata 21 

carnerosana 3, 

15, 16, 18, 19, 24, 37, 38, 71, 77, 
78; pis. 1, 2. 

chaparral 32 

coloma 43 

confinis 29 30 

constricta 3, 

15, 17, 53-55, 56, 58, 59, 62, 64, 
66, 68, 71, 73-75, 81, 87, 88; pis. 
34,D, 35,D, 36,C, 47. 
dwarf form 60 



96 



AGR. MONOGRAPH 17, U. S. DEPT. OF AGRICULTURE 



Page 
Yucca — Con. 

draconis var. arborescens 30 

elata 1-3, 

6-8, 10, 11, 13, 17, 19, 20, 24, 
27, 37, 38, 40, 45, 52-58, 59-65, 
66, 68, 70-76, 83, 84, 85-92; pis. 
37, A, B, 49, 50, 64, 65, 70-72. 

dwarf form 60 

faxoniana 3, 

6, 8, 10, 15, 16, 19, 20, 24, 28, 
37, 70, 71, 73, 77, 87, 89, 90, 92; 
pis. 3, 56. 

filamentosa 66 

gilbertiana 3, 

16, 17, 44-47, 52, 54, 55, 65, 72, 
73, 81, 82, 90; pis. 26, 27, 29,A. 

glauca 2, 

3, 5-8, 10, 11, 15, 17, 28, 37, 
43-45, 47-49, 50, 52, 54-59, 62-66, 
68, 70-77, 81-83, 84-88, 90, 91; 
pis. 30, 34,/i, 35,/i 16,A. 

alliance 55-59; 

pis. 31-33, 34,£, 35, 36, 63, 66. 

var. garneyi 48 

var. stricta 48 

gloriosa 66 

gramhiifolia 33 

Great Plains 47 

barrimaniae 43-47, 54, 55 

var. gilbertiana 44 

hoary 20 

hybrids, apparent: 

baccata-arizonica 29, 30, 73; pi. 14 

blue form 29 

carnerosana-torreyi 18 

elata-glauca alliance 62-65, 

74; pis. 37, C-F, 38-43. 

faxoniana-torreyi 20 

gilbertiana-neomexicana 47; 

pis. 28, 29, B, C. 

glauca-angustissima 56-58; 

pi. 36, D-F. 

-baileyi 56-58 

-constricta 56-59, 75; pi. 66 

-elata 56-59 

reverchoni-rupicola 42, 43 

schidigera-baccata var. vespertina. 25, 

30. 
thompsoniana-reverchoni ....40, 74, 75 

torreyi-baccata 23, 24 

treculeana-torreyi 21, 22 

whipplei-whipplei var. 

caespitosa 36, 81 

intermedia 56-58, 74 

var. ramosa 58, 62 

kanabensis 62, 64, 65 

longiflora 21 

louisianensis 55 

macrocarpa 11, 19, 22, 24 

Mohave 24 

mohavensis 24 

moth 12, 13, 65-69 

mountain 20 

navajoa 51 



Page 
Yucca — Con. 

neomexicana 3, 

16, 17, 43, 44 t 45-47, 50, 54, 55, 
59, 65, 66, 68, 71, 73, 81, 82, 

84, 85, 87, 88, 90, 91; pis. 24, 
25, 29, D, E, 61. 

newberryi 33, 37 

New Mexico 43 

pallida 41 

-palm 30 

peninsularis 36 

puberula 26 

Quixote 32 

radiosa 59,61 

reverchoni 3, 

11, 15, 17, 24, 40, 41-43, 59, 66, 

68, 71, 73-75, 81, 82, 85, 87, 88; 

pis. 23, 68, B. 
rostrata 3, 

11, 15, 17, 18, 38, 39, 40, 65, 71, 

76, 78, 80, 85; pi. 20. 

f. integra 38 

rupicola 3, 

4, 10, 15-17, 40, 41, 42, 43, 59, 
66, 68, 70, 71, 73, 81, 82, 85, 87, 
88; pi. 22. 

edentata 41 

tortifolia 40 

Salt Lake Desert 44 

San Angelo 41 

schidigera 3, 

4, 6, 8, 11, 17, 24-26, 28, 29, 37, 
68, 70, 71, 74, 77, 78, 79, 85, 87, 
89, 91, 92; pis. 8, 9, 57. 

blue form 24, 25 

schottii 3, 

4, 16, 17, 20, 24-27, 71, 73-75, 

77, 87, 89, 92; pis. 4, 5, 52. 

standleyi 49-51, 55, 58 

tenuistyla 54, 55 

thompsoniana 3, 

11, 15, 17, 24, 39, 40, 42, 43, 
65, 66, 68, 71, 73-76, 80, 81, 

85, 87, 88, 92; pis. 21, 68, A, 69. 

thornberi 29, 30, 73, 92 

Torrey 20 

torreyi 3, 

4, 8, 10, 15, 17-21, 22-24, 25, 26, 
28, 29, 38, 40, 59, 70, 71, 73-76, 
77, 81, 85, 87, 89-92; pis. 7, 58. 

blue form 22, 23,73 

f. parviftora 22, 23 

Trans-Pecos 39 

treculeana 3, 

15, 17, 21, 22, 23-26, 28, 59, 90; 
pi. 6. 

var. canaliculata 21 

treleasei 26 

utahensis 61-64 

verdiensis 62-64 

whipplei 2, 

3, 17, 25, 32-37, 66, 61, 70, 71, 
73-76, 80, 81, 85, 88, 90; pis. 
19, 53, A, 55. 



INDEX 



97 



Yucca whipplei — Con. 

subsp. caespitosa 54 

subsp. intermedia 34, 35 

subsp. parishii 33, 35 

subsp. percursa 35 

subsp. typica 33, 35 



Page, Page 

Yucca whipplei — Con. 

var. caespitosa 32, 

34, 35-37, 73, 75, 76, 80, 81, 85. 

var. intermedia 33, 34, 35, 36 

var. parishii 33, 35, 73, 75, 76, 80, 85 

var. percursa .... 33, 35, 36, 76, 81, 85 

Yucceae 71 



GOVERNMENT PRI 



IG OFFICE: 1953 — 210994 



Agriculture Monograph 17, U. S. Department of* Agriculture 



Plate 1 




Yucca carnerosana (Trel.) McKelvey: 45 miles south of Marathon, Tex., on road to 
Black Gap ranch. Photographed, March 29, 1944. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 1 




Yucca earner os ana (Trel.) McKelvey: On Black Gap ranch, approximately 60 miles 
southeast of Marathon, Tex. Plant with five trunklike stems and five basal sprouts. 
Photographed, March 30, 1944. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 3 




Yucca faxoniana (Trel.) Sarg. : 9 x /2 miles southwest of Chispa, Tex. Inflorescence 
immature. Photographed, April 4, 1944. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 4 




Yucca schottii Engelm. : 1 mile east of Patagonia, Ariz. Plants with several shoots 
and fresh leaves covering entire trunklike stems. Photographed, August 3> 1943. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 5 



. . ■ ■ - • - 



■ J ' >■ 




Yucca schottii Engelm. : 15 miles northwest of Cloverdale, N. Mex. Plants with few 
shoots and fresh leaves limited to apical crown. Photographed, April 12, 1945. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 6 




Yucca treculeana Carr. : 8 miles north of Crystal City, Tex. 
Photographed, March 27, 1944. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 7 




Yucca torreyi Shaf.: 3 miles east of State College, N. Mex. 
Photographed, June 30, 1942. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 8 




Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 9 




Yucca schidigera Roezl.: Same locality as plate 8. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 10 








1 4% 


o 


JM' 


n 


§ 


"2 






m < 


8 


<*» 


>< 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 11 




Yucca baccata Torr. : Salt River Canyon, about 15 miles northeast of Globe, Ariz. 
Photographed, July 12, 1942. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 12 




Yucca baccata var. vespertina McKelvey: 6 miles southwest of Peach Springs, Ariz. 
Photographed, July 14, 1942. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 13 




Yucca baccata var. vespertina McKelvey: Near Mule Springs, northern end of 
Lanfair Valley, Mojave Desert, Calif. Photographed, May 5, 1943. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 14 





. mt>m? s 



? Yucca baccata X ? Y. arizonica: l l / 2 miles northeast of Douglas, Ariz. 
Photographed, May 12, 1945. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 15 




Yucca brevifolia Engelm. : 15 miles east of Lancaster, Calif. An exceptionally fine 
plant. Photographed by Ernest Braunton, July 9, 1925. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 16 




Yucca brevifolia var. jaegeriana McKelvey: South slope of While Hills, 3 miles 
east of entrance to Cyclopic Mine, Northern Mohave County, Ariz. Photographed, 
October 14, 1943. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 17 




♦ ' 



' 



Yucca brevifolia var. jaegeriana McKelvey: 15 miles south of Windmill Station on 
road to Cima, near Kessler Spring, Mojave Desert, Calif. Undoubtedly the largest 
concentration of yucca in United States. Photographed, May 4, 1943. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 18 




ft 

< 



ft 



Agriculture Monograph 17, U. S. Department of Agriculti 



Plate 19 




Yucca whipplei Torr.: About 7 miles west of Beaumont, Calif. 
Photographed, April 24, 1948. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 20 



& ( 4 






«du 




















■■■; " 






Yucca rostrata Engelm. : Near Black Gap Ranch, about 55 miles southeast of 
Marathon, Tex. Photographed, March 30, 1944. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 21 




Yucca thompsoniana Trel.: 12 miles east of Fort Stockton, Tex. 
Photographed, May 30, 1943. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 22 




Yucca rupicola Scheele: 3 miles north of Mountain Home, Tex. 
Photographed, May 21, 1945. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 23 




Yucca reverchoni Trel. : 4 miles east of Barnhart, Tex. Photographed, May 30, 1943. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 24 




Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 25 




Yucca neomexicana Woot. & Standi.: 8 miles southeast of Kenton, Okla. 
Photographed, October 21, 1943. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 26 



a>\*" ->| %* 




Yucca gilbertiana (Trel.) Rydb.: 37 miles west of Delta, House Range, Utah. 
Photographed, June 11, 1945. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 27 




Yucca gilbertiana (Trel.) Rydb.: 47 miles west of Delta, House Range, Utah. 
Photographed, June 11, 1945. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 28 



- 4 






$ 




. ? Yucca gilbertiana X ' ¥"• neomexicana: Clark Valley about 10 miles northeast of 
Price, Utah. Photographed, June 6, 1947. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 29 




Yucca gilbertiana, Y. neomexicana, and apparent hybrids: Leaf specimens of (A) 
Y. gilbertiana, House Range, Utah; (B) apparent hybrid, Beaver, Utah; (C) 
apparent hybrid, Helper, Utah; (D) Y. neomexicana between Sapinero and 
Gunnison, Colo., and (£) Y. neomexicana north of Moses, N. Mex. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 30 



U 



*1 




Yucca glauca Nutt. : Near Toonerville, Colo. Photographed, July 1, 1947, by 
Orville A. Parsons, Soil Conservation Service. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 31 










Yucca glauca alliance ? hybrid: Near Duran, N. Mex. Photographed, June 13, 1943. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 32 




Yucca glauca alliance ? hybrid: U. S. Department Agriculture Field Station, 
Tucumcari, N. Mex. Racemes not more than 35 cm. long. Photographed, June 
3, 1944. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 3: 




Yucca glauca alliance ? hybrid: Same locality as plate 32. Panicles broad, rather 
ellipsoidal; scape long. Photographed, June 5, 1944. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 34 




Capsules of Yucca glauca alliance: A, Y. glauca, Hadley, Tex.; B, Y. baileyi, near 
Holbrook, Ariz.; C, Y. angustissima, 30 miles east of Peach Springs, Ariz.; D, 
Y. constricta, Sterling City, Tex.; and E, variations in small field near Tucumcari, 
N. Mex. 



Agriculture Monograph 17, U. S. Department or" Agriculture Plate 35 



9 99.999 



A 



C D 



#rt? tot 



• 



f ft tttt 



Seed of Yucca glauca alliance: A, Y. glauca, Hadley, Tex.; B, Y. baileyi, Holbrook, 
Ariz.; C, Y. angustissima, 30 miles of Peach Springs, Ariz.; D, Y. constrict a, 
Sterling City, Tex.; E, variations in small field near Tucumcari, N. Mex.; and F, 
variations near Rowe, N. Mex. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 36 




Yucca glauca alliance: Pistils and stamens of (A) Y. glauca, Hadley, Tex.; (B) 
Y. baileyi, 20 miles north of Gallup, N. Mex. ; (C) Y. constricta, 10 miles south 
of Uvalde, Tex.; (D, E, F) apparent Y. glauca-Y. angustissima hybrids — 
(D) Tucumcari, N. Mex., (£) 26 miles southwest of Santa Fe, N. Mex., and 
(F) 12 miles west of Albuquerque, N. Mex. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 37 




Yucca elata and ? Y. elata X . ? Y. glauca alliance: Pistils and stamens of (A) Y. 
elata, near Las Cruces, N. Mex.; (JB) Y. elata, White Sands National Monument, 
N. Mex.; (C, D) ? Y. elata X - p Y. glauca — C, Hot Springs, N. Mex.; and D, 
22 miles north of Socorro, N. Mex.; (£, F) ? Y. elata X ? Y. angustissima — E, 
Zion National Park, Utah, and F, 10 miles north of St. George, Utah. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 38 



38 

s 



^ 






II A 

.4 











, ? Yucca elata X « p ¥"• glauca alliance: 42 miles south of Socorro, N. Mex. 
Photographed, June 11, 1944. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 39 




? Yucca elata X « p Y. glauca alliance: 22 miles north of Socorro, N. Mex. 
Photographed, June 11, 1944. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 40 




? Yucca elata X ? glauca alliance: Same locality as plate 39. Scape thin; 
panicle branched at base. Photographed, June 11, 1944. 



Agriculture Monograph 17, U. S. Department of Agriculture 



V, 





Plate 41 










? Yucca elata X •' V. Rlauca alliance: Same locality as plate 39, Scape thick, 
panicle ellipsoidal, late flowering. Photographed, June 11, 1944. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 42 




? Yucca elata X - ? Y. glauca alliance seedling: 72 months old, grown at Riverside, 
' Calif! from seed collected near Grant, N. Mex. Raceform inflorescence. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 43 




? Yucca elata X ? Y- glauca alliance: Sister of seedling shown in plate 42, 
Foliaceous, panicled inflorescence. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 44 




Yucca baileyi Woot. & Standi.: 5 miles east of Winslow, Ariz. 
Photographed, May 26, 1944. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 45 




Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 46 




Yucca angustissima Engelm.: 1 mile north of Paulden, Ariz. 
Photographed, July 14, 1942. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 47 



& 




Yucca constricta Buckl.: 16 miles north of Junction, Tex. 
Photographed, April 7, 1944. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 48 




? Yucca elata X ? Y> angustissima: About 10 miles south St. George, Utah. 
Photographed by R. H. Peebles, May 9, 1940. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 49 




^•TBJfe^'i^?^ 



Yucca elata Engelm. : 5 miles northwest of Separ, N. Mex. An extremely dense 
concentration. Photographed, June 11, 1942. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 50 




Yucca elata Engelm.: 5 miles southwest of Orogrande, N. Mex. Typical sand dune 
plant. Photographed, May 15, 1945. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 51 






^*^a & 



™&*m£±*j^i&&$§£ 



^.^^^^sm^^mgd^m^ 




? Yucca elata X ? Y. angustisshna: 10 miles south of Cedar City, Utah. 
Photographed, June 13, 1945. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 52 




Yucca schottii seedlings: 6 to 34 weeks old, grown at Riverside, Calif., from seed 
collected near Patagonia, Ariz. Development of fibrous root system. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 53 




Yucca wbipplei {A) and Y. brevifolia var. jaegeriana (B) seedlings: 6 months old, 
grown at Riverside, Calif., from seed collected near Beaumont, Calif., and Chloride, 
Ariz., respectively. Note bulblike stem of Y. wbipplei. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 54 




Yucca baccata (A) and Y. arizonica (B) seedlings : 43 months old, grown at River- 
side, Calif. ; from seed collected near Mountain Park, N. Mex., and Nogales, Ariz., 
respectively. Development of sprouts from rhizomatous extensions. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 55 




Yucca whipplei seedlings: 51-month-old sisters from plant in Cajon Pass, Calif. 
A, a solitary segregate; B, a caespitose segregate; and C, a hybrid segregate. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 56 




Yucca jaxoniana: Base of stem of old plant, 9 miles south of Chispa, Tex. Typical 
fibrous root system of flat base plants. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 57 




Yucca schidigera: Base of old, fallen stem, 7 miles southwest of Kingman, Ariz. 
Typical fibrous root system of flat base plants. Sprouts from active margin of stem- 
base. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 58 




Yucca torreyi: Base of old stem near State College, N. Mex. Typical fibrous root 
system of rounded base plants. Sprouts slightly removed from old stem. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 59 




Yucca baccata: Rhizomatous and aerial stems of old plant 10 miles north of Las 

Cruces, N. Mex. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 60 




Yucca brevijolia var. jaegeriana: Base of fallen plant, 12 miles north of Cima, Calif. 
Agriculture Monograph 17, U. S. Department of Agriculture Plate 61 




Yucca brevijolia forma berbertii: Stem, rhizomes, and sprouts of an old plant near 
Twentynine Palms, Calif. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 62 




Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 63 




Yucca glauca alliance seedlings: 15 months old, from seed collected near Grant, 
N. Mex. Development of lateral rhizome system. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 64 




Yucca elata seedlings: 7 months old, grown in the United States Soil Conservation 
Service Nursery, Tucson, Ariz. Development of vertical rhizome system. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 65 




Yucca elata seedlings: 15 months old, grown at Riverside, Calif, from seed collected 
in White Sands National Monument, N. Mex. Development of vertical rhizome 
system. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 66 




? Yucca glauca X ? Y. constricta 20 miles west of Ozona, Tex. Lateral rhizome 

system. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 67 




Yucca neomexicana-. 35 miles north of Clayton, N. Mex. Lateral rhizome system. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 68 





'V.V 



Yucca 
sprout 



thompsoniana (^4) and Y. reverchoni (B) : A, Rhizomatous base of young 
near Bakersfield, Tex.; B, rhizomes of old leaf head, near San Angelo, Tex. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 69 




in ! 






* 




Yucca thompsoniana: 28 miles south of Marathon, Tex. Rhizome base of old plant. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 70 




Yucca elata: United States Jornada Range Experiment Station, Las Cruces, N. Mex. 

Vertical rhizome system. 



Agriculture Monograph 17, U. S. Department of Agriculture 



Plate 71 




Yucca elata: Shoot with adventitious roots, with old leaves at base of shoot. White 
Sands National Monument, N. Mex. 



Agriculture Monograph 17, U. S, Department of Agriculture Plate 72 





1022220786 



MIONAL AGRICULTURAL LIBRARY 



022220787 



NATIONAL AGRICULTURAL LIBRARY